The present disclosure provides an automatic image annotation method and apparatus, a device, and a medium. The method comprises: acquiring a plurality of sample images to be annotated, extracting image features of network representation layers on the basis of a convolutional neural network (CNN) feature extraction model, and constructing first KNN relationship graphs; inputting the first KNN relationship graph of a sample image to be annotated having a node density lower than an empirical value into a pre-trained GCN-V node affinity prediction model, to predict affinities between nodes in the KNN relationship graph; cutting edges between the nodes according to the affinities, to enable nodes of a same cluster to belong to a same type of samples, and pruning to obtain a second KNN relationship graph of said sample image; inputting the second KNN relationship graph of said sample image into a pre-trained GCN-C edge connection strength prediction model, to predict edge connection strength between clusters; and extracting the clusters having the edge connection strength greater than a threshold to form a cluster of said sample image.
G06V 10/762 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant le regroupement, p.ex. de visages similaires sur les réseaux sociaux
The present disclosure relates to an image recognition model training method and apparatus, a computer device, and a storage medium. The present disclosure can obtain, according to actual conditions and by means of simulation, training samples meeting a corresponding global non-independence-and-identical-distribution degree value, and using same for training, thus improving the training accuracy while ensuring the privacy security. The method comprises: acquiring from a master node a full training sample image set and a global non-independence-and-identical-distribution degree value, the full training sample image set comprising a plurality of kinds of recognition objects; determining a local target recognition object, and calculating the initial number of each kind of recognition objects in the full training sample image set; on the basis of a preset data construction rule, and according to the global non-independence-and-identical-distribution degree value, performing calculating to obtain the target number of each kind of target recognition objects; according to the target number of each kind of target recognition objects, constructing a target training sample image set; and using the target training sample image set to train an image recognition model to obtain a trained image recognition model.
G06V 10/774 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant l’intégration et la réduction de données, p.ex. analyse en composantes principales [PCA] ou analyse en composantes indépendantes [ ICA] ou cartes auto-organisatrices [SOM]; Séparation aveugle de source méthodes de Bootstrap, p.ex. "bagging” ou “boosting”
3.
GRAPH EMBEDDING-BASED RELATIONAL GRAPH KEY PERSONNEL ANALYSIS METHOD AND SYSTEM
Disclosed in the present disclosure is a graph embedding-based relational graph key personnel analysis method and system. The method comprises the following steps: constructing a character relationship graph on the basis of social media data; analyzing each node in the character relationship graph by means of a graph embedding algorithm to obtain an embedding vector of each node; generating key node seeds of the character relationship graph according to a pre-correlation index; and according to the embedding vector of each node, analyzing the key node seeds by means of a clustering algorithm to identify a key personnel node. In the present disclosure, topological properties of the relationship graph are fully utilized, learnability is achieved, and there is no need to manually set a parameter value or specify a calculation rule for degree gain, so that an adverse effect of unreasonable setting of a manual rule is eliminated; meanwhile, the whole graph is calculated, and the isomorphism and heterogeneity of the nodes are integrated, so that the obtained analysis result of key personnel is more accurate.
Provided are a robot joint braking device (100) and a robot joint (300). The robot joint braking device comprises: a rotating shaft (4), the rotating shaft being mounted on a robot joint body (200); a bolt assembly (1), the bolt assembly (1) comprising a support (12), a magnetic pushing member (10) connected to the support (12), and an elastic member (14) and a bolt (13), wherein the bolt (13) is displaced in the axial direction of the bolt (13) by means of a magnetic force of the magnetic pushing member (10) and an elastic force of the elastic member (14); and a braking assembly (2), the braking assembly comprising a friction column (21) and a brake disk (22), wherein the brake disk (22) is in interference connection with the friction column (21), an overlapping area exists in a projection of the brake disk (22) and a projection of the bolt (13) in the displacement direction of the bolt (13), and the braking assembly (2) controls the rotating shaft (4) to stop by means of braking the brake disk (22) in an interference and clamping manner in a displacement process of the bolt (13).
B25J 19/00 - Accessoires adaptés aux manipulateurs, p.ex. pour contrôler, pour observer; Dispositifs de sécurité combinés avec les manipulateurs ou spécialement conçus pour être utilisés en association avec ces manipulateurs
F16D 65/14 - Mécanismes d'actionnement pour freins; Moyens pour amorcer l'opération de freinage à une position prédéterminée
5.
IDENTIFICATION METHOD AND SYSTEM FOR QUICK GATHERING BEHAVIOR OF CROWD, AND DEVICE AND MEDIUM
Disclosed in the present disclosure are an identification method and system for a quick gathering behavior of a crowd, and a device and a medium. The identification method comprises: acquiring continuous frames of images, and sequentially performing head detection on each frame of image on the basis of a head detection model; performing clustering analysis on a gathering region of a full frame of image according to a head detection result, calculating the number of heads in each clustered gathering region, and calculating a congestion coefficient corresponding to each frame of image; when there is a target frame of image, a congestion coefficient of which is greater than a preset sparsity coefficient, taking the target frame of image as a starting point, calculating and updating congestion coefficients and the maximum number of persons in the gathering region of all the frames of images within a preset time period; and determining whether there is a frame of image within the preset time period, the congestion coefficient and the maximum number of persons in the gathering region of which both exceed corresponding preset threshold values thereof, and if so, then outputting a result of quick gathering of a crowd. The present disclosure solves the problem of difficulty in terms of feature expression caused by the serious lack of human body contours when a crowd is dense; moreover, the adaptability of an identification method is also improved.
The present application relates to vehicle type identification, and in particular relates to a vehicle type identification method, a device, and a system. The vehicle type identification method comprises: acquiring a vehicle front image, a vehicle body image, and a vehicle back image of a vehicle; obtaining width, height, and length data of the vehicle, the width, height, and length data comprising first width-height data, second width-height data and third width-height data; obtaining vehicle type data of the vehicle, the vehicle type data comprising vehicle type information and axle information; and combining the width, height, and length data and the vehicle type data to output vehicle type identification data. After the vehicle front image, the vehicle body image, and the vehicle back image are obtained, license plate evaluation is carried out, and then a first proportionality coefficient is obtained according to a license plate image size and a license plate physical size; subsequently, accurate first width-height data of the front of the vehicle, second width-height data of the back of the vehicle, and third width-height data of the vehicle body are obtained, and the vehicle type and the axle type are also accurately identified; data of a vehicle that has undergone vehicle type identification can be accurately obtained, and identification accuracy can be improved.
The present disclosure relates to an intelligent network model, and particularly relates to a federated learning method and system based on a blockchain. The federated learning method based on a blockchain comprises: a plurality of consensus nodes training a federated model according to sample data, so as to respectively obtain corresponding first gradient values and send same to a coordination terminal; the coordination terminal aggregating the received plurality of first gradient values, so as to obtain second gradient values and send same to the respective consensus nodes, such that each of the consensus nodes performs parameter updating on the federated model on the basis of the second gradient values; and determining whether the updated federated model meets a termination condition, if so, terminating training, and if not, performing training again. Compared with a traditional serial federated learning method, time can be significantly saved; moreover, the consensus nodes all equally perform recording and calculation, without relying on the integrity of an initial node, such that the security during the training process can be improved.
Disclosed are a block chain transaction data retrieval system, a method, a device, and a storage medium. The retrieval system comprises: a block chain monitoring node, which is used for synchronizing transaction data and a block on a block chain, and sending, using blocks as units, data obtained via monitoring to a data processing module; the data processing module, which is used for receiving block data of different zone blocks, and after separately processing different pieces of block data, caching same in a memory database; a data storage module, which is used for performing classification and screening at regular times on data stored in the memory database in the data processing module according to structure of a graph database, and writing a processed structure into the graph database; and a data query module, which is used for carrying out real-time retrieval on data written into the graph database by using a query language of the graph database and according to a received query instruction. The present disclosure can quickly perform real-time relationship retrieval on a transaction on a block chain, and real-time transaction tracing and tracking is implemented.
A multi-modal currency recognition system, a method, a storage medium and a terminal, belonging to the technical field of currency recognition. The system and the method use texture sensing and open set recognition technology; recognition capability of depth features is improved by means of multi-modal feature fusion, and recognition performance is improved by means of a texture sensing multi-currency recognition algorithm. An open set recognition algorithm can achieve recognition of new currencies and new denominations of banknote by means of a small number of samples, overall development costs are low, and recognition accuracy is high.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
An intelligent design method for a personalized discount coupon, and an electronic apparatus and a storage medium. The method comprises: acquiring a video stream that is captured by a camera (S101); executing a target tracking task on a pedestrian and a commodity in the video stream, so as to obtain a tracking result (S102); calculating commodity sales condition features according to the tracking result, wherein the commodity sales condition features at least comprise the number of times the commodity has been viewed, the number of times the commodity has been inquired about, the number of times the commodity has been taken, and the number of times the commodity has been placed into a shopping cart (S103); training a regression model by using the commodity sales condition features, so as to obtain a sales prediction model (S104); predicting, by using the sales prediction model, the sales volume of the commodity under a given discount (S105); and on the basis of a predicted sales volume, generating a discount coupon under the condition of profit maximization (S106). The method can assist a merchant in rationally controlling marketing costs and can also increase profits, without violating the privacy of users, such that the problems of high marketing costs, a low sales volume and low profits caused by a previous method for issuing discount coupons according to personal experience are solved.
A visualization processing system and method for unstructured data, and a storage medium and a terminal, which belong to the technical field of data visualization processing. The system comprises a data visualization processing module, a data engine module and a data source module, which are in two-way communication, wherein the data visualization processing module comprises an application template library, a chart library, a condition constructor and a drawing component, and is used for receiving and rendering data results, which are from the data engine module and are subjected to unified analysis and processing, so as to realize data presentation meeting diversified requirements. In the method, by means of a unified data processing solution, the problem of it being impossible to directly visualize unstructured data in the past is solved, the problem of an application being required to be redeployed after a visualized chart is generated is also solved, and service personnel can easily realize visualization analysis of mass data.
The present disclosure provides a large file encryption and decryption system, a method, a storage medium, and a device, in the technical field of data protection; the system comprises a sender block chain node, a receiver block chain node, an encryption module, a decryption module, a public key of a sender, a private key of the sender, a public key of a receiver, a private key of the receiver, and a symmetric key. The method comprises: the sender encrypts plaintext and the receiver decrypts ciphertext; an identity of a file owner is determined by means of a block chain private key, and a high-performance encryption mechanism is provided to improve encryption performance; by means of secondary encryption and decryption, it is ensured that only a specified receiver can decrypt the file, ensuring the security of the file contents.
The present disclosure relates to a rapid authentication method based on finger vein biometric recognition technology, and a device and a medium. The rapid authentication method comprises: distributing finger vein templates in a registration template library to a front computing node or a plurality of parallel computing nodes according to a preset distribution rule; and in response to a recognition request that is sent by a vein terminal, sending a collected vein image to the corresponding front computing node or parallel computing nodes to perform feature matching processing, so as to output a final authentication result. In the present disclosure, the biometric recognition efficiency is improved by means of a combination of parallel computation, front recognition and user mode self-learning.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
14.
FINGER VEIN RECOGNITION METHOD AND SYSTEM BASED ON MULTI-PARTY COOPERATION, AND STORAGE MEDIUM
The present invention relates to the technical field of biological recognition, and provides a finger vein recognition method and system based on multi-party cooperation, and a storage medium. In a recognition algorithm, a lightweight finger vein recognition algorithm based on an attention mechanism is introduced, and the attention mechanism can pay more attention to a network structure to suppress other interference factors, so that the algorithm is high in recognition accuracy and low in consumed computing power, and thus, the performance of the algorithm is improved; and by using a multi-user cooperative training framework, secret leakage can be avoided, cooperative common training can be realized by means of a cooperation center, and the robustness of a model can be improved.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
A data association analysis method and apparatus, and a computer device and a storage medium, which relate to the technical field of data analysis, and can realize fast association analysis of heterogeneous data, thereby improving the data association analysis efficiency. The method comprises: determining at least two types of heterogeneous data sources; extracting metadata information of heterogeneous data from the heterogeneous data sources, forming an intermediate table that includes the metadata information of the heterogeneous data, and storing the intermediate table in a memory database; taking, as target metadata information, metadata information in the memory database that is associated with a data component during an association analysis process; according to the target metadata information, acquiring, from the corresponding heterogeneous data sources, target heterogeneous data that corresponds to the target metadata information, and storing the target heterogeneous data in the memory database; and running the association analysis process to analyze the target heterogeneous data in the memory database, so as to obtain a data association analysis result.
An image processing method and apparatus, a device, and a storage medium, relating to the technical field of computer processing. The method mainly comprises: acquiring a first connected graph obtained by clustering an image set (S201); predicting, according to an order of descending betweenness centrality measures of connection edges comprised in the first connected graph, whether the connection edges are incorrect connection edges (S202); if a target connection edge which is an incorrect connection edge is predicted, obtaining a reconstructed connected graph of the first connected graph, the reconstructed connected graph being obtained by performing reconstruction according to an edge-excluded graph of the first connected graph and the edge-excluded graph comprising a graph in the first connected graph excluding the target connection edge (S203); and determining a second connected graph of the image set according to the first connected graph and the reconstructed connected graph (S204). The present invention ensures image clustering accuracy while saving computing resources and reducing time costs.
A video synopsis method, system and device. The method comprises: obtaining a current-frame image and obtaining a background image on the basis of the current-frame image; calculating on the basis of each first pixel point of the background image to obtain a first LBP feature vector; calculating on the basis of each second pixel point of the current-frame image to obtain a second LBP feature vector; comparing the first LBP feature vector with the second LBP feature vector to obtain a static pixel point mask; obtaining a first foreground image mask on the basis of the background image and the current-frame image; resetting the first foreground image mask on the basis of the static pixel point mask to obtain a second foreground image mask; performing connected component analysis on the second foreground image mask to obtain a current-frame detection targe queue; performing moving target tracking on the current-frame detection target queue to output the positions and IDs of current-frame detection targets; and generating, on the basis of the positions and IDs of the current-frame detection targets, a video file after synopsis.
G06F 16/738 - Présentation des résultats des requêtes
G06F 16/783 - Recherche de données caractérisée par l’utilisation de métadonnées, p.ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement utilisant des métadonnées provenant automatiquement du contenu
G06T 7/194 - Découpage; Détection de bords impliquant une segmentation premier plan-arrière-plan
18.
SHORT-TERM TRAFFIC FLOW PREDICTION METHOD UNDER COMPLEX ROAD NETWORK, STORAGE MEDIUM, AND SYSTEM
Provided are a short-term traffic flow prediction method under a complex road network, a storage medium, and a system. The method comprises: collecting traffic flow data of a plurality of road segments; performing data cleaning and time slicing on the traffic flow of the plurality of road segments to obtain multi-dimensional traffic flow data; on the basis of short-term traffic flow model training, training the traffic flow data using a long short-term memory network; and on the basis of a trained short-term traffic flow model, performing short-term traffic flow prediction on the plurality of road segments under the complex road network using the long short-term memory network. The method, under the complex road network, can directly perform, by means of a long short-term memory network classifier, short-term traffic flow prediction on N road segments to be predicted, and is not limited to traffic flow prediction of a single upstream or downstream road segment, and the processing method is simple and effective; the method has strong learning and memory capacities and a good prediction effect for short-term traffic flow training and prediction, and has strong algorithm robustness, real-time performance, and strong applicability.
G06Q 10/04 - Prévision ou optimisation spécialement adaptées à des fins administratives ou de gestion, p. ex. programmation linéaire ou "problème d’optimisation des stocks"
The present invention provides an attention mechanism-based system and method for detecting a feature in a target, and a storage medium, and relates to the field of intelligent security. The system comprises a semantic extraction module, an attention map module and a detection module, and a classification sub-module is responsible for carrying out global attribute classification on a target and for supervising the training of an attention sub-module; the attention sub-module is responsible for constructing an attention map; and the detection module comprises an anchor frame filter layer, a target detection layer and a parsing layer, the anchor frame filter layer performing data filtering on a received result of the attention map module, sending the result to the target detection layer and the parsing layer for detection and analysis, and outputting a detection result. In the present invention, a multi-task learning method based on a deep convolutional network is employed, and attention learning and a single-scale detection mechanism are introduced, a feature in a target is detected and positioned, and global attributes of the target are classified and recognized, which solves the problem that in training stages of conventional solutions, sample distribution is unbalanced and the requirement for computing power is high due to multiple anchor frames and multiple scales, thereby improving detection efficiency and precision.
A small network model pruning method and system. The method comprises: searching channels according to a preset rule to determine a channel to be pruned; determining indexes of reserved channels; and performing weight reconstruction. The small network model pruning method realizes filter-level network model compression and acceleration.
The present invention belongs to pattern recognition technology in the field of transportation. Provided are a method for recognizing the color of a vehicle body of a vehicle, and a storage medium and a terminal. The method comprises: extracting mask regions of a vehicle body, a vehicle window and a license plate of a vehicle; acquiring a vehicle-body candidate region image to be subjected to color recognition; and filling the vehicle-body candidate region image and recognizing the color. By means of the solution, the color of a vehicle body can be quickly recognized, and the recognition accuracy can be improved.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
A banknote quality test method and system based on a multi-spectral image, and a storage medium. The banknote quality test method comprises: collecting a multi-spectral image of a banknote to be tested, and pre-processing the image to obtain a plurality of sub-regions; extracting a feature parameter of each sub-region, performing feature fusion on feature parameters of the sub-regions, and then importing same into a trained regression model to obtain a degree of confidence; and determining the quality of said banknote by means of combining the degree of confidence and a regional voting policy. Feature training is performed in advance on banknotes without abnormal features by using a regression model, and when a banknote to be tested needs to be tested, a feature vector of said banknote is imported into the trained regression model, such that graffiti and smudginess detection in any complex background can be completed, thereby achieving the aim of quickly testing the quality of a banknote.
A file scanning method, comprising: receiving a target image, extracting contour features, first filtering processing, overlapping processing, second filtering processing, calculating distance feature values, calculating one-dimensional head-up feature values, facing determination, and rotation correction. The described method achieves the orderly storage of the target image, manual intervention is not needed for the entire process, the position of a file in a scanning apparatus does not need to be checked when a user scans the file, a file to be scanned can be placed into the scanning apparatus in any direction, and manual intervention is not needed for a later stage, thereby greatly increasing the digital processing efficiency of a physical information image of existing files.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06T 3/60 - Rotation d'une image entière ou d'une partie d'image
24.
DEPOSIT BANKNOTE DISTRIBUTION SCHEDULING METHOD AND SYSTEM, AND STORAGE MEDIUM
Disclosed in the present invention are a deposit banknote distribution scheduling method and system, and a storage medium. The method comprises: according to banknote box types, banknote box states, acceptable denominations, remaining spaces, wear degrees and slot positions of banknote boxes, configuring priority information of the banknote boxes to obtain a banknote box total queue containing the priority information of the banknote boxes; and when a user stores a banknote, controlling the banknote to be stored in a banknote box according to the priority information in the banknote box total queue. According to the present invention, the banknote box to which the banknote is stored is an optimal banknote box, so than the banknote deposit in banknote boxes is optimized, the deposit success rate is improved, the probabilities of banknote blockage and device failure are reduced, and thus the client deposit experience is improved. At the same time, the banknote stored by a client can be recycled as much as possible, so that the number of times of cleaning a machine and adding banknotes by a bank worker is reduced, and the personnel cost and capital cost of a bank are reduced.
G07D 11/24 - Gestion de la réserve en papiers de valeur
G07D 11/18 - Dérivation vers des trajets ou récipients différents
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
25.
CELLOPHANE DEFECT RECOGNITION METHOD, SYSTEM AND APPARATUS, AND STORAGE MEDIUM
Disclosed in the present invention are a cellophane defect recognition method, system and apparatus, and a storage medium. The cellophane defect recognition method specifically comprises: step S1, collecting cellophane surface images to establish test set data; step S2, importing the test set data into a semantic segmentation network model based on an optimized UNET network for semantic segmentation; and step S3: acquiring an output signal of the semantic segmentation network model, and post-processing the output signal to obtain a defect recognition result. According to the present invention, a conventional image processing method and a semantic segmentation network are combined as the basis for automatic classification, thereby effectively improving the robustness of cellophane defect recognition compared with conventional image processing and analysis; moreover, the semantic segmentation model uses the optimized UNET network, such that a cellophane defect can be quickly detected, the time required for actual detection and analysis is reduced, the efficiency and accuracy of cellophane defect detection are improved, and detection costs are reduced.
A deep learning-based fingerprint texture extraction method, specifically comprising: step S1: obtaining a fingerprint image, and preprocessing the fingerprint image to enable the size of the fingerprint image to meet an input requirement of a fingerprint texture extraction model; step S2: inputting the preprocessed fingerprint image into the fingerprint texture extraction model to output a texture map; and step S3: cutting the output texture map to obtain a texture map having the size consistent with the size of the original fingerprint image. Also disclosed are a deep learning-based fingerprint texture extraction system and apparatus, and a storage medium. By means of the method, fingerprint textures in various conditions can be well extracted, and the accuracy for fingerprint identification can be greatly improved.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
27.
IDENTITY CARD NUMBER DETECTION METHOD AND APPARATUS, AND READABLE STORAGE MEDIUM AND TERMINAL
Provided are an identity card number detection method and apparatus, and a readable storage medium and a terminal. The method comprises: firstly, positioning the position of an identity card by using a target detection algorithm, and extracting an image from a detection box; performing size normalization on the image; selecting a mask of an appropriate scale to perform a morphological operation on the image, such that identity card number areas are connected; adaptively selecting a threshold value to realize binarization; acquiring a straight line by using Hough; selecting an image main direction and performing rotation; and extracting a connected area, i.e. an identity card number area, the aspect ratio of which meets requirements. By means of the detection solution, there is no need to perform inclination correction on an identity card, and the extraction of identity card information in a complicated photographing scene can be improved, thereby greatly improving the recognition rate; and the present invention can be widely applied to the fields of security, finance, etc.
Provided are a certificate category increase and decrease detection method and apparatus, a readable storage medium, and a terminal. The method comprises: firstly, storing standard pictures of certificates of various categories in a memory as registered pictures; secondly, detecting certificates to be detected, so as to obtain an input picture, and performing image processing on a picture to be input; and finally, comparing the processed picture with the registered pictures, determining, by means of a similarity, a category to which the detected picture belongs, so as to perform quick and efficient screening and category determination on newly-added certificates. By means of the detection solution, the accuracy and efficiency of determining a category of a newly-added certificate in a complex photographing scenario can be improved, and the detection solution can be widely applied in the fields of security, finance, etc.
A collaborative learning method that dynamically updates a central party, a storage medium, a terminal and a system. The system comprises a data provider determination module, a performance evaluation module, a preference module which, according to a performance evaluation, independently selects the best data provider in a network to act as a central server and run a central program; a communication transmission module; and a dynamic updating learning module which, according to an initial status, determines whether to access a collaborative learning task or update the central server; if yes, the collaborative learning task is started and the risk for the central server is predicted; if not, the the central server is accessed for updating, and it is determined whether the collaborative learning task is continued, recovered or stopped until the task ends. Compared to the prior art, the present solution provides a method in which when a central server crashes, an optimal data provider is rapidly selected from among participating training tasks as the central server, which allows for rapid linking so that the collaborative learning can operate continuously.
Disclosed is a method for processing the high concurrency of a big data annotation service. The method of the present invention comprises: a distributed buffer conducts a registration service to a coordination module; the coordinate module reads/writes picture index information having a timestamp in an index library, and sends the picture index information to the distributed buffer; the distributed buffer feeds back the concurrent access amount to the coordination module; the coordination module adjusts and allocates the picture index information to the distributed buffer according to the concurrent access amount; a client reads the picture index information of the distributed buffer and downloads an annotation task picture; the client submits annotation information to an annotation platform server, and the annotation platform server updates the annotation state of index information of the annotation task picture. The present invention solves the problem of high concurrent read/write by adding a memory buffer, and provides a high-throughput concurrent annotation service method, which overcomes the disadvantages of the prior art of high concurrent read/write of a data annotation service due to the fact that a policy for a high concurrent readers-writer lock is not provided during data annotation.
Provided are a method, storage medium, and system for central party selection in a collaborative deep learning task; the solution comprises: on the basis of performance score values of collaborators within an initial term, performing initial selection of a central party and using same for deep learning model aggregation; during performance of the task, performing central party update selection on the basis of the performance score values in the surviving nodes in the current term network and using the updated central party to continue the model aggregation task; this provides a stable central party for a learning method or model aggregation, and, in case of a central party abnormality, enables fast selection of an optimal data provider to participate in a collaborative learning task as a central party; quick linking is possible so that the model training process can continue to run, and the invention can be broadly applied in economic, cultural, educational, medical, and public administration industries.
The present invention discloses a finger vein image restoration method based on partial convolution and mask updating. Said method comprises: detecting an anomalous region (noise region) of a damaged finger vein image, and marking the anomalous region; generating the damaged finger vein image with the anomalous region mark into a corresponding original mask image; and inputting the damaged finger vein image and the original mask image into a pre-trained irregular hole image restoration model, wherein the irregular hole image restoration model performs, according to the original mask image, a feature encoding and decoding operation on a non-anomalous region on the damaged finger vein image by means of partial convolution and mask updating, and outputting a restored image of the damaged finger vein image. For the finger vein image restored by the present invention, the finger vein texture is clear, the transition of the original edge of the noise region is natural, the restoration process is simple, the effect is obvious, and the success rate and the speed of finger vein image restoration are improved.
Disclosed in the present invention are a data set division method and system in a federated learning scene. The method comprises: determining whether data distributions of raw data provided by various federated learning participants are consistent; performing training using the raw data provided by the various federated learning participants and model test data to form a federated classification model; inputting the raw data belonging to the various federated learning participants into the federated classification model, the federated classification model outputting a probability of model input data being the model test data; and according to prediction probabilities in a descending order, selecting a designated number of model input data as a verification set provided by the federated learning participant to which the data belongs and used for verifying the model performance, and the remaining model input data as a training set provided by the federated learning participant to which the data belongs and used for training the model. The present invention can find, from the data provided by various federated learning participants, data samples most similar to the data distribution of the test data set to serve as a verification set for model training.
Provided are a threshold setting method based on a video quality inspection scenario, a storage medium, and a system. Two thresholds, a qualitative and a quantitative, are set for quality inspection function points: the qualitative threshold is used to determine which rules, determined from a logic level, a current service judgment is following, such as same frame detection, and said threshold needs to clearly define what is a same frame; the quantitative threshold is used to determine a tightness of a current algorithm, and said threshold is used to determine a video-based service processing result and not an algorithm processing result of a single image. In the present solution, an algorithm threshold setting method is provided for two typical function points, "same frame detection" and "information verification", in a video quality inspection scenario, thus solving a logic connection problem between algorithm analysis and service judgment.
A bankcard tilt correction (BTC)-based detection method and apparatus, a readable storage medium, and a terminal. By using BTC technology in combination with deep learning technology and a conventional image processing method, the advantages of the two are fully integrated, and for a wide variety of user input images having complex scenes, high-accuracy and high-robustness certificate segmentation and correction results can be obtained, thereby providing a foundation for subsequent certificate detection, classification, and information extraction, and improving the application range of certificate recognition; the present invention can be widely applied in the fields of security, finance, and the like.
The present application relates to a finger vein-based identity identification method and apparatus, a computer device, and a storage medium. Said method comprises: acquiring branch information and bifurcation point information in vein skeleton diagrams corresponding to two finger vein images to be matched; respectively performing matching on the branch information and the bifurcation point information of the two vein skeleton diagrams to obtain multiple groups of matched bifurcation point pairs; acquiring coordinate offset amounts of each group of matched bifurcation point pairs, and obtaining multiple groups of target bifurcation point pairs by screening the groups of matched bifurcation point pairs according to the coordinate offset amounts; acquiring a similarity of each group of target bifurcation point pairs, and acquiring branch lengths of all branches in the vein skeleton diagrams; and according to the similarity of each group of target bifurcation point pairs and the branch lengths of all branches in the vein skeleton diagrams, obtaining a similarity of the two finger vein images, and performing identity identification according to the similarity. The accuracy of a result of calculation of a similarity of two finger vein images can be improved by using said method.
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
A turntable-type medium processing device, comprising: a base component (18), a coin dispensing passage (7), a turntable (19), a coin push-up pin (20), and a coin push-up pin elastic piece (2). The turntable (19) is horizontally mounted on the base component (18); the turntable (19) is provided with circular holes through which a medium can pass; the coin dispensing passage (7) is provided at the side of the base component (18) corresponding to the turntable (19); the coin push-up pin (20) is mounted at the side of the base component (18) corresponding to the coin dispensing passage (7) and at a position corresponding to the circular holes; the coin push-up pin elastic piece (2) is connected to the bottom of the coin push-up pin (20). The medium is pushed up out of the coin dispensing channel (7) by means of the coin push-up pin (20); by means of the coin push-up pin (20) and the coin push-up pin elastic piece (2) at the bottom, when the turntable (19) rotates counterclockwise, the coin push-up pin (20) can make elastic motion and thus enter the next circular hole from the bottom of the turntable, thereby pushing up a medium in the next circular hole where the state of the medium is changed from horizontal to inclined. The technical problem that a medium is prone to be stuck when a turntable is placed horizontally is solved, and the space utilization is improved.
G07D 9/00 - Comptage de pièces de monnaie; Manipulation de pièces de monnaie non prévue dans les autres groupes de la présente sous-classe
G07D 9/06 - Dispositifs pour empiler ou disposer de toute autre façon sur un support des pièces de monnaie, p.ex. plateau à orifices utilisables pour compter des pièces de monnaie
G07F 11/00 - Appareils déclenchés par des pièces de monnaie pour distribuer ou délivrer d'une façon analogue, des articles d'une façon discontinue
38.
FACE VERIFICATION METHOD AND APPARATUS BASED ON TRIPLET LOSS, AND COMPUTER DEVICE AND STORAGE MEDIUM
The present invention relates to a face verification method and apparatus based on Triplet Loss, and a computer device and a storage medium. The method comprises: based on a face verification request, acquiring a certificate photograph and a scenario photograph of a person; respectively performing face detection, key point positioning and image pre-processing on the scenario photograph and the certificate photograph, so as to obtain a scenario face image corresponding to the scenario photograph and a certificate face image corresponding to the certificate photograph; inputting the scenario face image and the certificate face image into a pre-trained convolutional neural network model used for face verification, and acquiring a first feature vector corresponding to the scenario face image and a second feature vector corresponding to the certificate face image, which feature vectors are output by the convolutional neural network model; calculating a cosine distance between the first feature vector and the second feature vector; and comparing the cosine distance with a pre-set threshold value, and determining a face verification result according to a comparison result. The method improves the reliability of face verification.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
A venous image acquisition device. A main body is provided with an opening for incidence of natural light. A second polarizer is provided at the opening. A hole is provided at a side surface of the main body for inserting an acquisition object. An infrared light source is provided within the main body and emits infrared light to illuminate the acquisition object inserted into the hole. An imaging assembly is provided within the main body, and receives infrared light transmitting through the acquisition object, so as to perform venous imaging. The first polarizer is provided within the main body, located between the imaging assembly and the hole, and polarizes the infrared light and first polarized light directed toward the imaging assembly, the first polarized light being polarized light generated by a second polarizer polarizing the natural light. The polarization directions of the first polarizer and the second polarizer form a pre-determined non-parallel angle. The invention solves the technical problems in which a user using a top-lit venous image acquisition apparatus cannot see the position of their own fingers during use, and a side-lit venous image acquisition apparatus causes overexposure of a venous image.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
40.
BLOCK CHAIN CONSENSUS METHOD, APPARATUS AND SYSTEM, AND IDENTIFICATION INFORMATION PROCESSING METHOD AND APPARATUS
The present application relates to a block chain consensus method. A client sends behavior information carrying a behavior request to a master node; the master node receives the behavior information, and sends an identification generation request carrying the behavior request to a first trustable device; the first trustable device generates identification information carrying a unique number of the behavior request, and sends the identification information to the master node; the master node broadcasts the identification information to slave nodes, implements the behavior request, and sends a generated implementation result to the client; the slave nodes receive the identification information, and then send an identification verification request carrying the identification information to a corresponding second trustable device; each second trustable device verifies validity of the unique number in the identification information; the slave nodes receive a validity result returned from the corresponding second trustable device, then implements the behavior request, and sends a generated implementation result to the client; the client determines a consensus result on the basis of the received implementation result. The solution provided by the present application can reduce network and hardware overheads.
Disclosed are a warning method, a device, and a terminal for failure risk of a station terminal, and a storage medium. The method comprises: acquiring current operation data of a station terminal, the current operation data comprising a current health parameter; extracting, from a pre-established failure feature knowledge base, a health parameter reference comprising a pre-determined threshold range of the current health parameter and failure information corresponding to the health parameter reference, performing computation on the extracted health parameter reference and the corresponding failure information by means of a pre-determined algorithm, and obtaining a current failure risk status of the station terminal, the failure risk status comprising a failure type and a probability of occurrence of the failure type; and if the probability of the occurrence of the failure type reaches a pre-determined failure warning threshold, sending a failure warning signal. Embodiments of the present invention are adopted to warn of high-potential failure risks of a station terminal, thereby improving timeliness of failure handling.
A medium aggregation and separation device, comprising: a box body; the box body comprises at least two stacking parts and at least one reversing block; at least two stacking part channels are formed between an entrance of the box body and the stacking parts; the reversing block is disposed at the fork of the stacking part channels, the reversing block being a component that rotates about a fixed rotary shaft; when the reversing block rotates to a preset position, one stacking part channel is in communication while the remaining stacking part channel is closed such that a medium enters into a corresponding stacking part by means of the entrance of the boxy body and one stacking part channel. By means of dividing the box body into at least to stacking parts, the present invention may, according to the type of a medium, further convey the medium into a corresponding stacking part by means of the steering of the reversing block, and may increase the type of stacked media manifold such that one box body may be used for at least two media, thereby expanding the adaptability of existing machine core modules to medium type, and solving the technical problem wherein traditional financial tools cannot simultaneously support multiple currencies.
A finger vein image collection equipment, device and method, comprising: a main body, a detection module, a near-infrared light source, a visible light source, an image collection module and a control module; a light-transmitting hole is provided on the main body; the detection module is arranged on the main body and is used for detecting whether an object to be identified is placed at the light-transmitting hole or not; the image collection module is arranged in the main body and is used for receiving, from a second side of the light-transmitting hole, a first image of the object to be identified after being irradiated by visible light and a second image of the object to be identified after being irradiated by near-infrared light; and the control module is further used for comparing whether the first image is consistent with the second image or not, and if not, determining the second image to be a real finger vein image and using the second image as a final collected finger vein image. The present invention solves the technical problem in an existing finger vein image collection method of a lawbreaker printing or copying a finger vein image onto white paper so as to feign a real finger vein image which is collected and passes identity verification.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
44.
THIN-SHEET MATERIAL FRONT AND BACK SURFACE ADJUSTMENT DEVICE
A thin-sheet material front and back surface adjustment device (100) comprises: a material input channel (1), comprising at least one pair of first conveying wheel groups (11) for conveying a thin-sheet material and an identification mechanism (12) for identifying the front and back surface of the conveyed thin-sheet material; a material conveying channel (3), used to convey the thin-sheet material downstream and comprising at least one pair of third conveying wheel groups (31) for conveying the thin-sheet material; and a material flipping roller (2), comprising a roller (21) formed by two opposite semicircular channel assemblies, wherein a conveying channel (22) for conveying the thin-sheet material is formed between the two semicircular channel assemblies. Rotation of the roller is controlled by a first driving mechanism. Two ends of the conveying channel are dynamically connected to the material input channel and the material conveying channel. A second conveying wheel group (23) for conveying the thin-sheet material and a first position sensor (24) for determining the position of the thin-sheet material are provided in the conveying channel. The adjustment device has a simple structure and is easy to control.
Disclosed are a user identity verification method, apparatus and system based on a blockchain. The user identity verification method comprises: acquiring a user identification code and user information ciphertext from a client; the user information ciphertext is generated by encrypting, using a current user private key, user information to be verified; sending the user identification code to a blockchain serving end so that the blockchain serving end acquires corresponding current user information and a current user public key; using the current user public key returned by the blockchain serving end to decrypt the user information ciphertext, and acquiring decrypted user information to be verified; when it is determined that the current user information and the decrypted user information to be verified are the same, generating a verification passed message; and the client triggering, according to the verification passed message, a refresh operation on the current user private key and the current user public key. The present invention can improve the security of a user identity verification process and the accuracy of a user identity verification result.
The present invention relates to an identity authentication method, comprising the following steps: obtaining an identity card image, a face visible light image, and a face near-infrared light image of a person to be authenticated; inputting the identity card image, the face visible light image, and the face near-infrared light image into a Triplets CNN model which is trained in advance, and extracting the convolution features of the identity card image, the face visible light image, and the face near-infrared light image to obtain corresponding feature vectors; calculating the similarity between any two of the identity card image, the face visible light image, and the face near-infrared light image separately according to the feature vectors; and determining the consistency among the identity card image, the face visible light image, and the face near-infrared light image according to the similarity, and outputting an identity authentication result. In the present invention, the Triplets and the deep convolutional neural network (CNN) model are effectively combined, so that the robustness of identity authentication can be effectively improved, and thus the accuracy of identity authentication is increased.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
Disclosed are a method and apparatus for dispensing a medium. The apparatus for dispensing a medium comprises a rack (1), a medium box ejection mechanism (3), a medium box cabinet (2), and a push-up mechanism (4), wherein a dispensing opening (14) is provided on the top of the rack (1), and a vertically arranged first channel (13) is provided in the interior of the rack, the top end of the first channel (13) being aligned with the dispensing opening (14); the medium box cabinet (2) is used for accommodating several vertically arranged medium boxes and has an ejection opening (20) in communication with the interior of the rack (1); the medium box ejection mechanism (3) is used for pushing each medium box in the medium box cabinet (2) toward the ejection opening (20) while keeping same in a vertical state; and the push-up mechanism (4) is used to transfer the current medium box, which is pushed to the ejection opening (20), upward to the dispensing opening (14) through the first channel (13). The apparatus for dispensing a medium can be applied to the scenario where a whole unit is located at a lower end and a medium-dispensing position is located at an upper end, and can push the medium box up and dispense same while keeping same in a vertical state.
The present invention relates to a communication system, method and apparatus for an embedded self-service terminal. The communication system comprises: an application layer and a hardware device operation layer. The application layer comprises an interface matching module, a first serialization module and a first communication module. The hardware device operation layer comprises a second communication module, a second serialization module and a core processing module. The interface matching module is used for acquiring an API interface of a target program, and matching the API interface to an SPI interface complying with the hardware device operation layer. The first serialization module is used for parsing memory data into byte stream data. The first communication module and the second communication module are used for transceiving the byte stream data between the application layer and the hardware device operation layer. The second serialization module is used for parsing the byte stream data into the memory data. The core processing module is used for generating, according to the memory data, a corresponding command to drive a hardware device. The present invention can realize cross-platform, cross-system and cross-platform communication control on an embedded self-service terminal.
H04L 29/06 - Commande de la communication; Traitement de la communication caractérisés par un protocole
H04L 29/08 - Procédure de commande de la transmission, p.ex. procédure de commande du niveau de la liaison
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
The present invention relates to a portable identity verification apparatus, comprising an upper base body, a lower base body and a connecting body, wherein the upper base body is rotatably connected to one end of the connecting body, and the lower base body is rotatably connected to the other end of the connecting body; a first camera and a display screen are provided on the upper base body, an identity reader is provided on the lower base body, and a controller is provided in the connecting body; and the controller is respectively electrically connected to the first camera, the display screen and the identity reader. The upper base body and the lower base body rotate in the direction towards the connecting body, such that the upper base body and the lower base body approach the connecting body, and after being folded, the apparatus is approximately cuboid, which is convenient for packaging and transportation; in addition, the apparatus can be used as a hand-held device, is convenient for use in places with high population mobility, thereby reducing requirements for packaging and transportation conditions, and is highly adaptable.
The present invention relates to a method and apparatus for verifying person and certificate. The method comprises: acquiring a certificate facial image in a certificate photograph, and collecting a natural light facial image of a user; inputting the certificate facial image into a pre-trained generative adversarial network, and obtaining a reconstructed facial image corresponding to the certificate facial image according to an output of the generative adversarial network, wherein the generative adversarial network is used for adding pre-set natural light attribute information to the input certificate facial image, and the resolution of the reconstructed facial image output thereby is higher than the resolution of the certificate facial image; and comparing the reconstructed facial image with the natural light facial image, and verifying a person and certificate according to a comparison result. According to the present invention, the accuracy of verifying a person and certificate can be effectively improved.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
Disclosed in the present invention is a banknote graffiti detection method. A watermark region of a banknote is used as the standard; gray-scale histograms of the watermark region on a reflection image and a transmission image are calculated respectively; then weighted summation is performed so as to obtain a weighted gray-scale histogram; mean value operation is performed according to the weighted gray-scale histogram so as to obtain a gray-scale background threshold; accumulative summation is performed on a gray-scale histogram of a non-watermark blank region in the reflection image and the gray-scale histogram of the watermark region so as to obtain a gray-scale histogram of the reflection image; proportion operation is performed on the basis of the gray-scale background threshold and the gray-scale histogram of the reflection image to obtain a graffiti index of the banknote to be detected. Thus, the problems of a complex and time-consuming algorithm in an existing banknote image graffiti detection method are resolved; the degree of graffiti of the banknote can be quantitatively described; the present invention is suitable for various different types of banknotes, and can achieve quick and effective effects.
A method and device for detecting a spliced banknote. Edge detection and enhancement processing are performed, by using a horizontal Sobel operator and a vertical Sobel operator, on an image of an area to be detected, to obtain a first image and a second image; projection processing is performed on the first image and the second image to obtain a horizontal grayscale projected curve and a vertical grayscale projected curve; whether the projection result of the horizontal grayscale projected curve and the vertical grayscale projected curve meets a requirement of a normal banknote is determined, so as to identify whether a banknote is a spliced banknote. The invention can accurately detect spliced banknotes and eliminates the need to perform binarization processing and Hough transform processing on a banknote image, reducing detection complexity of spliced banknotes, and improving detection efficiency.
A method and system for detecting the authenticity of a banknote. The method for detecting the authenticity of a banknote comprises the following steps: step 1, image acquisition and correction: acquiring an image of a banknote, and rotating and cropping the image according to measured vertex information and angle information, to find a valid banknote image region; step 2, key feature point positioning: positioning the key feature points of the image of the banknote to be detected, and calculating the size feature values of the image of the banknote to be detected according to the relative coordinates of the positioned key feature points; 3, authenticity determination: comparing the size feature values of the image of the banknote to be detected with standard size feature values corresponding to a preset standard banknote; if the error between the size feature values of the image of the banknote to be detected and the standard size feature values exceeds a threshold, determining the banknote to be detected to be an abnormal banknote; and if the error does not exceed the threshold, determining the banknote to be detected to be a normal banknote. According to the present method and system for detecting the authenticity of a banknote, the global error of the measurement of the image of the banknote is localized by means of the relative coordinates of a plurality of key feature points, and a plurality of absolute errors are relativized, so that the purpose of weakening errors to improve recognition accuracy is achieved.
Disclosed are a coin roll dispensing device and coin roll distribution apparatus. The coin roll dispensing device comprises a coin storage box, a coin dispensing mechanism and an anti-arch-bridge mechanism. The coin storage box is used for storing coin rolls. The coin storage box has an open front end. A coin temporary storage channel is arranged at the front end of the coin storage box. The coin dispensing mechanism is arranged at an outlet of the coin temporary storage channel and used for dispensing coin rolls to a coin bag. The anti-arch-bridge mechanism is arranged on the coin storage box at an inlet of the coin temporary storage channel and is used for eliminating the arch bridge phenomena formed due to the stacked coin rolls at the inlet of the coin temporary storage channel. With the coin roll dispensing device and the coin roll distribution apparatus, coin rolls can be stored in a multi-layered stacking manner, such that more coin rolls can be stored, and reliable coin dispensing is achieved.
The present application provides a quality score calculation method for a finger vein image. The method comprises: acquiring a normalized image and a binary image of a finger vein image of which the quality is to be determined; acquiring a boundary point and an intermediate point of each finger vein in the binary image; then acquiring gray-scale values of the boundary point and the intermediate point in the normalized image, and calculating an average gray-scale difference between the intermediate point and the boundary point; creating an average gray-scale difference array of the intermediate point and the boundary point according to the average gray-scale difference; calculating points that are greater than a preset threshold in the average gray-scale difference array; finally summing the points with different preset thresholds to obtain cumulative sums of the points with different preset thresholds; and normalizing the cumulative sums of the points to obtain a quality score of the finger vein image. Hence, by means of the present solution, a high-quality finger vein image is obtained, high accuracy is achieved, then identity authentication is performed, and high authentication efficiency is achieved.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
56.
CARD DOWNWARD-DISPENSING DEVICE AND AUTOMATIC TICKET/CARD DISPENSER
A card downward-dispensing device and an automatic ticket/card dispenser. The card downward-dispensing device comprises a ticket/card box seat frame, a card friction wheel assembly and a movable rear rolling wheel assembly provided within the ticket/card box seat frame. The card friction wheel assembly is provided at a position in the middle of the bottom of the ticket/card box seat frame, close to the front side, and comprises a card friction wheel for applying friction to a card so as to generate a frictional force to push the card out of the ticket/card box. The movable rear rolling wheel assembly is provided on the rear side of the bottom of the ticket/card box seat frame, and comprises a rear rolling wheel which can move up and down. The bottom surface of the ticket/card box is provided with an opening, and the card friction wheel and the rear rolling wheel protrude from the opening to enter the bottom surface of the ticket/card box and keep in contact with cards within the ticket/card box. As the rear rolling wheel has the capability of adaptively moving up and down, in the case of there being many cards, the cards can meet the need for slope climbing, and in the case of there being few cards, the slope climbing process is cancelled, solving the problem that the card outputting success rate is low.
A paper money authenticity identification technique, being, in particular, a paper money authenticity identification method which performs self-verification by means of a unique double prefix number of paper money and a system for implementing the method; first, a paper money image signal is acquired; then, image pre-processing is carried out, comprising bilinear interpolation and binarization; then, characters are divided equally into four regions, black dot percent is counted respectively, and similarity comparison of horizontal and vertical black dot percent is carried out; X-direction and Y-direction projection is carried out horizontally and vertically respectively, and projection energy is counted; lastly, by combining the similarity of black dot percent and the similarity of projection energy, final determination of coupling degree is performed to determine whether or not horizontal and vertical characters are consistent, thereby identification of counterfeit paper money is effectively carried out. The accuracy of the present method is high, while algorithm implementation is simple and the amount of computation is light, thus the present invention is suitable for embedded financial machinery and may effectively improve the authenticity identification performance of financial machinery.
G07D 7/0047 - Vérification spécialement adaptée à la détermination de l’identité ou de l’authenticité de papiers de valeur ou pour discriminer ceux qui sont inacceptables, p.ex. qui ne correspondent pas à une monnaie en utilisant des éléments de sécurité numériques, p.ex. des informations codées sur un filet ou une bande magnétique en utilisant des codes de contrôle, p.ex. des chiffres codés dérivés d’un numéro de série et de la valeur nominale
An image processing method, the method comprising: acquiring the image data of a current object to be detected which has been collected by a first image collection module; in the case where the first image collection module does not emit light and a second image collection module adjacent to the first image collection module emits light, predetermining an interference signal from the second image collection module to an image collected by the first image collection module; and according to the interference signal, removing interference from the data of each pixel point among the acquired image data to generate actual image data. In the case where image collection modules are placed at adjacent positions, the present application eliminates the influence of the light source of one module on the image collection module on the other side, eliminating the mutual interference, such that the acquired image can stably and reliably restore the real image of a bank note, thereby ensuring the ability to identify counterfeit bank notes. In addition, the present invention further provides an image processing device and an image processing system which have the described technical advantages.
Provided are a sheet flattening device (110) in the field of financial self-service devices and a financial self-service device with the sheet flattening device (110) installed therein. The sheet flattening device (110) comprises: an outer channel transmission unit, an inner channel transmission unit and a power drive unit. The power drive unit is used for providing power to the flattening device (110). A sheet flattening channel is formed between the outer channel transmission unit and the inner channel transmission unit. The sheet flattening channel comprises at least one bending portion, the inner channel transmission unit is provided with an edge on the bending portion, and when a sheet passes through the bending portion of the flattening channel, the outer channel transmission unit provides pressure enabling the sheet to be closely adhered to the edge. A rolled medium, after passing through the flattening channel, is flattened by means of a bending structure, so as to achieve the technical effect of being flattened.
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
60.
METHOD AND DEVICE FOR DETECTING SEE-THROUGH REGISTER ANTI-COUNTERFEITING CHARACTERISTICS
Provided are a method and device for detecting see-through register anti-counterfeiting characteristics. The method comprises: acquiring a first detection image and a second detection image of an object to be detected; extracting anti-counterfeiting characteristic information about the first detection image to generate a first anti-counterfeiting characteristic image; extracting anti-counterfeiting characteristic information about the second detection image and performing mirror transformation to generate a second anti-counterfeiting characteristic image; fusing the first anti-counterfeiting characteristic image and the second anti-counterfeiting characteristic image to generate a fused anti-counterfeiting characteristic image; and matching the fused anti-counterfeiting characteristic image with a standard anti-counterfeiting characteristic image in a pre-stored template, and according to the matching result, determining the authenticity of the object to be detected. The present application only requires the collection of visible light reflection images of front and back faces, and does not require the collection of a visible light transmission image, thereby effectively reducing the cost of a device where a see-through register anti-counterfeiting performance detection is ensured, and does not require the application of a complex algorithm of image splicing technology, thereby reducing the performance requirement for a digital signal processor.
The invention discloses a bank note one-dimensional signal detection method and device. The method comprises: converting bank note one-dimensional signal data into a two-dimensional gray value; preprocessing the two-dimensional gray value; and detecting the two-dimensional gray value through a two-dimensional data processing method, so that it can be detected whether an adhesive tape is bonded on the bank note, the authenticity and face value of the bank note. The present invention solves the problems in the prior art that, in the existing one-dimensional signal detection, abnormal samples are difficult to be distinguished, and normal samples are falsely denied, and bank notes bonded with adhesive taps are difficult to be identified accurately.
An automatic ticket vending and dispensing machine, comprising a machine body frame (10), a human-machine interaction module, a ticket processing module (13), a control module, and a communication module. The human-machine interaction module receives a ticket dispensing instruction input by a user and delivers the same to the control module. The control module, upon receiving the ticket dispensing instruction, provides a two-dimensional code with an identifier of the automatic ticket vending and dispensing machine to the user via the human-machine interaction module. The control module acquires, via the communication module, order information of the user pushed by an order system in a network. The human-machine interaction module acquires order information indicating that the user has confirmed ticket output, and, via the control module, delivers the order information indicating that the user has confirmed ticket output to the ticket processing module (13), to complete ticket output. The automatic ticket vending and dispensing machine of the present invention achieves Internet ordering and on-site ticket collection functions without employing a two-dimensional code identifying module.
Disclosed are an image compensation method and device. By determining valid windows and valid signals, calculating a background value of an original signal, further calculating a compensation value of the original signal and finally determining and outputting a compensation signal, the method and device solve the technical problem in the prior art that system cost and operation difficulty are increased because large memory is needed for saving compensation coefficients and an additional temperature sensor is needed. The method in the embodiments of the present invention comprises: determining valid windows according to a preset parameter calculation window that is determined on the basis of an obtained original signal of a scanned image, and calculating valid signals corresponding to the valid windows according to the determined valid windows; calculating a background value corresponding to the original signal according to the calculated valid signals; calculating a background reference value of the original signal according to the background value and the preset parameter calculation window; and determining a compensation signal corresponding to the original signal according to the background reference value and the background value.
Disclosed is a calibration method for a finger vein identification apparatus. The method comprises the following steps: using an infrared image sensor of a finger vein identification apparatus to collect L respective finger vein images of N persons in a current state, wherein N≥2 and L≥1 (S101); calculating Gaussian distribution parameters of section pixel grayscale values of the N*L finger vein images, wherein the section pixel grayscale values of each of the finger vein images comply with a Gaussian distribution (S102); by means of formula (1), calculating a Gamma mapping coefficient of the infrared image sensor, wherein (μs, σs) and γs are pre-set standard values of the finger vein identification apparatus; (μ, σ) is a Gaussian distribution parameter obtained in the previous steps; and γ is a Gamma mapping coefficient of the infrared image sensor (S103); and according to the Gamma mapping coefficient obtained by means of calculation, adjusting a mapping relationship between a pixel input value f and a pixel output value f1 obtained by the infrared image sensor in the current state, and the finger vein identification apparatus then identifying the adjusted pixel output value f1, so as to obtain an identification result (S104). The calibration method can effectively reduce the influence of factors, such as the climate, temperature and illumination of different regions, on the identification success rate of the finger vein identification apparatus.
A cross-platform drive test method, a device and a system, the method comprises: reading a test instruction sent by a test host, the test instruction comprising test data and a module identification code (S1); sending the test data to a hardware emulation module corresponding to the module identification code, such that the hardware emulation module processes the test data (S2). With this method, drive test efficiency and stability can be improved.
A lock mechanism, a banknote box and financial self-service equipment, for locking and unlocking a mobile carrier on a fixed carrier. The lock mechanism comprises a lock mounting substrate (3) mounted on the fixed carrier; a lock hook plate (35) mounted on the lock mounting substrate (3) in a sliding manner; and a locking fastener mounted on the mobile carrier and corresponding to the lock hook plate (35). The locking or unlocking of the mobile carrier is implemented by sliding of the lock hook plate (35). The lock mechanism further comprises a lock-out plate (36) and a locking driving sheet (22); the non-end portion of the lock-out plate (36) is provided with a rotating pivot rotationally connected with the lock mounting substrate (3); the inner-end portion of the lock-out plate (36) is in linkage connection with the locking driving sheet (22) mounted on the fixed carrier; the outer-end portion of the lock-out plate (36) pushes against or is separated from the lock hook plate (35) under the driving of the locking driving sheet (22); when the outer-end portion of the lock-out plate (36) pushes against the lock hook plate (35), the lock hook plate (35) and the locking fastener are locked. The risk that a lock hook plate of the existing lock mechanisms is detached from the locking fastener under the impact of an external force is overcome.
The present application provides a finger vein acquisition method and device. The acquisition method comprises: a finger sensing unit receives a first trigger instruction and sends a generated sensing signal to a control unit; after receiving the sensing signal, the control unit controls a first light source and a second light source in multi-angle light sources to be separately turned on or off; an image acquisition unit obtains at least two finger vein images when different light sources are turned on or off; an image fusion unit splices and fuses the finger vein images acquired by the image acquisition unit; and an image storage unit stores the spliced finger vein image obtained after fusion. Hence, by acquiring finger vein images from different angles and performing feature matching, fusion, and splicing, the finger vein acquisition method and device provided by the present invention can effectively solve the problem of rotation of a finger during a vein acquisition process.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
A finger vein recognition method comprises: capturing finger vein images at multiple angles (S3); after stitching the finger vein images captured at multiple angles into a target matching image (S4), capturing an image of a region of interest (S1); and matching the image of the region of interest and the target matching image, a successful match indicating that finger vein recognition is successful (S2). The above method captures finger vein images at different angles, and performs feature matching to realize recognition of a finger vein, thereby exhibiting high accuracy. Applying the invention in identity verification achieves high verification efficiency.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
69.
CARD ISSUING AND REPLACEMENT APPARATUS FOR SELF-SERVICE BANKING SYSTEM, METHOD FOR SAME, AND SELF-SERVICE BANKING TERMINAL
A card issuing and replacement apparatus for a self-service banking system, method for same, and self-service banking terminal. The apparatus comprises: a card reader (1) used to read and store account information of a first card, write the account information of the first card into a second card, and then eject the second card; a card issuing device (2) dispensing a stored second card; a temporary card storage device (3) for temporarily storing the first card; a card recycling device (4) for recycling the first card and/or the second card; and a card transporting device (5) capable of switching sequentially between the following operation modes including a first operation mode in which the first card is taken from the card reader (1) and transported to the temporary card storage device (3), a second operation mode in which the second card is taken from the card issuing device (3) and transported to the card reader (1), and a third operation mode in which the first card is taken from the temporary card storage device (3) and transported to the card recycling device (4).
A pasted paper money detection method and device. The method comprises: collecting a paper money image (S1); performing coarse positioning on an area probably for pasting in the paper money image to acquire a coarse positioning area image (S2); performing grey level histogram statistics on the coarse positioning area image, performing grading on the coarse positioning area image in accordance with a statistical result according to a grey level value interval, and dividing the coarse positioning area image into a plurality of grey level images (S3); respectively performing binarization processing on each of the grey level images, and performing envelopment fitting on the binarized images to acquire an image envelope of each of the grey level images (S4); and performing straight line detection on the image envelope, and if it is detected that the image envelope includes two straight lines not belonging to paper money itself, marking the paper money corresponding to the paper money image as pasted paper money (S5). According to an image feature of pasted paper money shown in a paper money image, with regard to patterns with different grey level conditions probably formed by a pasted tape, the detection rate of the pasted paper money can be effectively improved, the solution is easy to realize, and the equipment costs are economized.
A note stacking method and system, and self-service banking apparatus. The note stacking method comprises: detecting, by means of a note detection device (11), a physical state of note units (S1); calculating, according to the physical state of the note units, an optimal note stacking parameter of the note units, and sending, by means of a delivery device (13), the note units into a note stacking device (14) (S2); and after the note units are sent into the note stacking device (14) and the optimal note stacking parameter is met, triggering the note stacking device (14) to stack the note units. The present invention enables dynamic adjustment of an optimal note stacking parameter of note units on the basis of a detected physical state of the note units, thus improving neatness of note stacking.
A sheet medium flattening apparatus, comprises a support frame (101), a first conveyor wheel set (104) and a flattening block (111); the support frame (101) is provided with a first passage plate (103) and a second passage plate (102), the gap between the first passage plate (103) and the second passage plate (102) being a conveyance passage; the first conveyor wheel set (104) is able to convey a sheet medium (200); the flattening block (111) is able to protrude into and retract from the conveyance passage through a through hole on the first passage plate (103); the second passage plate (102) is provided with a groove cooperating with the flattening block (111), and the flattening block (111) is able to protrude, when the first conveyor wheel set (104) conveys the sheet medium (200), into the conveyance passage and push the sheet medium (200) into the groove so as to make the sheet medium (200) flat again. The flattening block in the sheet medium flattening apparatus is able to push the sheet medium into the groove in the second passage plate so as to make the arc-shaped sheet medium flat again, allowing a user to conveniently take out the sheet medium. Further included is a financial self-service device using the sheet medium flattening apparatus, being able to flatten an arc-shaped banknote, and prevent the banknote from being stuck during transmission in the financial self-service device.
Disclosed are an offline identity authentication method and an apparatus which are used to solve the problem that the prior art relies on a human face database provided by the Ministry of Public Security and that it is difficult to identify a three-way consistency of a chip avatar, a document surface avatar and a certificate holder image. The method comprises: acquiring two or more images for identity authentication to construct a to-be-authenticated multivariate image group (101); extracting a cascaded PCA convolution feature for each image in the to-be-authenticated multivariate image group to obtain a corresponding eigenvector (102); merging the image information of the to-be-authenticated multivariate image group according to the eigenvector and a fractional fusion strategy with a supervised signal to obtain a corresponding merged vector (103); and sending the merged vector to an SVM classifier which has been preliminarily trained to determine the consistency of the image in the to-be-authenticated multivariate image group, so as to obtain an identity authentication result (104).
G06K 9/64 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques utilisant des comparaisons ou corrélations simultanées de signaux images avec une pluralité de références, p.ex. matrice de résistances
A card reader system and method. An identity reading module is disposed at a financial service terminal. An identity security module is disposed at a server. A plurality of financial service terminal comprise identity reading modules capable of performing data exchange with the identity security module disposed in the server, thereby implementing a verification process verifying an identity document. In comparison to a conventional method in which an identity reading module and an identity security module are connected and disposed in a financial service terminal, the embodiment employs only one identity security module at a financial institution network node having a plurality of financial service terminals, and is only required to maintain only one identity security module, simplifying a process of maintaining an identity security module, and reducing costs.
A card container (102) realizing a security interlock is used to store hard plastic cards (901). The card container (102) is installed on a card issuing device. The plastic cards (901) are delivered by means of a card delivery channel (102a) of the card container (102). The card container (102) comprises a main body frame (1), a lock cylinder (6), a gate mechanism (2), and a locking mechanism (3). The lock cylinder (6), the gate mechanism (2), and the locking mechanism (3) are attached to the main body frame (1). The locking mechanism (3) fits with and moves in a linked manner with the lock cylinder (6) to selectively extend out of or retract back to the card container (102) within a preset angle range of the lock cylinder (6) with rotation of a spindle of the lock cylinder (6). The gate mechanism (2) and the locking mechanism (3) are linked. When the locking mechanism (3) extends out of the card container (102) to lock the card container (102) to the card distribution device, the gate mechanism (2) moves upwards to open the card delivery channel (102a). When the locking mechanism (3) retracts back to the card container (102) to release the card container (102) from the card distribution device, the gate mechanism (2) moves downwards to close the card delivery channel (102a).
G07B 5/00 - APPAREILS À DÉBITER DES TICKETS; APPAREILS À ENREGISTRER LES PRIX DES BILLETS; DISPOSITIONS OU APPAREILS POUR ENCAISSER LE PRIX DES BILLETS OU LES DROITS D’ENTRÉE OU DE PÉAGE EN UN OU PLUSIEURS POINTS DE CONTRÔLE; APPAREILS À AFFRANCHIR - Détails des machines à débiter les tickets ou dispositifs accessoires pour celles-ci
G07B 1/00 - Machines pour imprimer et débiter des tickets
G06K 13/08 - Alimentation en cartes ou délivrance de cartes
76.
METHOD AND DEVICE FOR ENCODING AND DECODING NEGOTIABLE INSTRUMENT INFORMATION, PROCESSING SYSTEM, AND FINANCIAL SELF-SERVICE EQUIPMENT
A method for encoding negotiable instrument information comprises: obtaining currency information, version information, and denomination information of a negotiable instrument; querying a preconfigured currency information association table according to the currency information and obtaining a currency code corresponding to the currency information; querying a preconfigured version information association table according the version information and obtaining a version code corresponding to the version information; accessing a preconfigured denomination information association table according to the denomination information and obtaining a denomination code matching the denomination information; and generating an information code of the negotiable instrument according to the currency code, the version code, and the denomination code. A decoding method, an encoding device, a decoding device, a processing system, and a financial self-service equipment are provided. The method can compatibly encode all new versions and new denominations of negotiable instruments, and each code table is stored in the hardware in the form of code data. The data cannot be changed manually, and thus there is no tamper risk.
G07D 11/00 - Dispositifs acceptant des pièces de monnaie; Dispositifs acceptant, distribuant, triant ou comptant des papiers de valeur
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
G06F 9/00 - Dispositions pour la commande par programme, p.ex. unités de commande
G06F 17/30 - Recherche documentaire; Structures de bases de données à cet effet
77.
DEVICE MANAGEMENT METHOD AND SYSTEM BASED ON ACTIVE TEMPLATE LIBRARY (ATL), AND FINANCIAL SELF-SERVICE DEVICE
A device management method and system based on an active template library (ATL), and a financial self-service device, for solving the problem that hardware device resources in existing ATM management systems cannot be shared. The method comprises: after receiving a communication connection request from an upper application for a hardware device, authorizing and verifying the upper application (A101); if the upper application passes the authorization and verification, establishing a communication connection between the upper application and a hardware device in order of priority, in the same single device management service process (A102); and in the same single device management service process, using an upper application with lower priority as a monitoring program, to monitor the communication between the current upper application with higher priority and the hardware device (A103). By means of the invention, two or more upper applications can share and access the same single hardware device by means of the same single device management service process, avoiding wasting device resources.
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
H04L 9/30 - Clé publique, c. à d. l'algorithme de chiffrement étant impossible à inverser par ordinateur et les clés de chiffrement des utilisateurs n'exigeant pas le secret
78.
THIN CARD-LIKE MEDIUM HANDLING DEVICE, AND SELF-SERVICE TICKETING MACHINE
A thin card-like medium handling device, and self-service ticketing machine. The device comprises: a support frame, and a stacking mechanism and a switching mechanism (9) mounted on the support frame. The stacking mechanism comprises an upper note-stacking assembly (1) and a lower note-stacking assembly (2), and the upper note-stacking assembly (1) and the lower note-stacking assembly (2) are tangent with respect to each other at a first end and openable at a second end. The switching mechanism (9) enables selectively opening or closing the second end of the upper note-stacking assembly (1) and the lower note-stacking assembly (2). The handling device and self-service ticketing machine have functions of transporting, stacking, aligning, recycling, etc., and have a simple structure and superior adaptability to a medium.
B65H 29/12 - Délivrance ou progression des articles à la sortie des machines; Progression des articles vers ou dans les piles au moyen du resserrement entre deux rubans ou courroies mobiles, ou entre deux jeux de rubans ou courroies
B65H 29/14 - Délivrance ou progression des articles à la sortie des machines; Progression des articles vers ou dans les piles au moyen du resserrement entre deux rubans ou courroies mobiles, ou entre deux jeux de rubans ou courroies et introduisant les articles dans une pile
G07B 3/00 - Machines pour débiter des tickets imprimés à l'avance
79.
SERIAL PORT MAPPING METHOD AND DEVICE BASED ON LINUX OR UNIX KERNEL OPERATING SYSTEM
A serial port mapping method based on a Linux or Unix kernel operating system. The method comprises the following steps: obtaining a first serial port list from a basic input/output system (BIOS), the first serial port list storing serial port information of at least one physical serial port of a panel (S101); obtaining a second serial port list from an operating system, the second serial port list storing serial port information of at least one logic serial port of the operating system (S102); and matching the physical serial port in the first serial port list with the logic serial port in the second serial port list according to the serial port information to generate a serial port mapping file of the first serial port list and the second serial port list (S103). The method and device can automatically generate a serial port mapping file for describing a mapping relationship between a physical serial port and a logic serial port without human interference, thereby improving the efficiency and accuracy of identification and matching.
Provided are an anti-skimming electromagnetic interference detection method and device. The detection method comprises: generating an original magnetic stripe signal, and acquiring a composite interfered magnetic stripe signal (S1), wherein the interfered magnetic stripe signal is formed by combining the original magnetic stripe signal and an interference signal from an electromagnetic interference source to undergo detection; performing decoding on the interfered magnetic stripe signal to obtain decoded parameters (S2); and determining whether original magnetic stripe parameters corresponding to the original magnetic stripe signal are consistent with the decoded parameters (S3), if so, a detection result indicates that the electromagnetic interference source cannot satisfy a requirement of a magnetic card reader with respect to the electromagnetic interference; otherwise, a detection result indicates that the electromagnetic interference source can satisfy a requirement of a magnetic card reader with respect to electromagnetic interference. The anti-skimming electromagnetic interference detection method and device of the present invention can accurately determine whether an electromagnetic interference source can satisfy a requirement of a magnetic card reader with respect to electromagnetic interference, thus enhancing the safety of the magnetic card reader and avoiding skimming of magnetic card information.
G01R 31/00 - Dispositions pour tester les propriétés électriques; Dispositions pour la localisation des pannes électriques; Dispositions pour tests électriques caractérisées par ce qui est testé, non prévues ailleurs
81.
PARALLEL HARDWARE ARCHITECTURE AND PARALLEL COMPUTING METHOD FOR FLOATING POINT MATRIX INVERSION
Disclosed in embodiments of the present invention is a parallel hardware architecture for floating point matrix inversion, for use in resolving the problem of fixed matrix dimensions when an existing matrix operation is implemented by hardware. The parallel hardware architecture for floating point matrix inversion in the embodiments of the present invention comprises: a matrix writing module, used for writing matrix data of an augmented matrix in a first memory and a second memory, the first memory and the second memory dynamically allocating storage spaces to the matrix data; an initial pivot positioning module, used for positioning an initial pivot row of the augmented matrix; a row switching module, used for switching the first row of the augmented matrix with the pivot row; a row elimination module, used for carrying out normalization and row elimination calculation, and positioning a next pivot row; and a triggering module, used for repeatedly and alternately triggering the row switching module and the row elimination module, and outputting the matrix data in the second memory until when the matrix to be inverted stored in the first memory becomes a unit matrix. The embodiments of the present invention also provide a parallel computing method for floating point matrix inversion.
Provided are a method and device for reliably verifying the identity of a financial card user. The method comprises the following steps: transmitting, to a financial card, response data required by the financial card, interacting with the financial card to acquire ciphertext response information comprising card number information and a first application ciphertext, and receiving transaction password information entered by a user; and transmitting, to a server of a card issuing party, the ciphertext response information, the transaction password information and the response data, interacting with the server of the card issuing party to obtain a first authentication result and first authentication data, transmitting the first authentication data to the financial card, obtaining a second authentication result after the financial card has performed verification on the basis of the first authentication data, and determining validity of the identity of the user according to the first authentication result and the second authentication result. The solution disclosed in an embodiment of the invention can verify the identity of a user and issuance of a financial card in a convenient, fast and reliable manner without needing an additional hardware device, thereby preventing security risk imposed by frauds, and ensuring information security.
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
83.
LINUX-BASED REGISTRY SERVICE SYSTEM AND METHOD, AND FINANCIAL SELF-SERVICE DEVICE
Disclosed is a Linux-based registry service system, comprising: a memory registry construction unit for loading a registry data file in an operating system, reading configuration table data already saved in the registry data file by means of a database interface, and constructing, in a memory, a memory registry; a registry operation unit for collecting node data from a registry editor or a user process, and transmitting the node data to a registry service unit; and the registry service unit for performing, according to an operation mode for the memory registry, a corresponding operation on a node corresponding to the node data in the memory registry. Also disclosed are a Linux-based registry service method and a financial self-service device. By performing, in a memory, an operation of node data in a memory registry of, a unified, efficient and secure registry configuration data operation can be provided for running all the software, and the requirement of data sharing between various processes is satisfied.
Disclosed is a hardware device access management method, comprising: when it is detected that a hardware device accesses a predetermined interface, a serving end establishing a connection with the hardware device, and creating a monitoring thread; when a connection request from a client is monitored, the monitoring thread creating a serving thread serving for the connection request; and the service thread receiving a protocol data packet sent from the client, invoking a corresponding function to establish a connection with the interface according to the protocol data packet, and serializing an output parameter of the function and then returning same to the client, so that the client de-serializes the output parameter of the function, then writes same into a function parameter table and then returns same to a corresponding upper-layer application by means of a driver. Also disclosed is a hardware device access management system, so that the upper-layer application can securely establish a connection with the hardware device in the case of having no root right.
An assembled note-issuing apparatus (100), and financial self-service apparatus. The assembled note-issuing apparatus comprises a base (10), and a note container (20), a paper delivery assembly (30), a printing assembly (40), and a temporary storage assembly (50) sequentially arranged on the base (10). The paper delivery assembly (30), the printing assembly (40), and the temporary storage assembly (50) are in communication with the note container (20) via a note channel, and the paper delivery assembly (30) delivers, via the note channel, a note in the note container (20) to the printing assembly (40) to print. The printing assembly (40) transports to the temporary storage assembly (50) the printed note for temporary storage. The assembled note-issuing apparatus (100) achieves functions of storage, delivery, printing, and temporary storage and recycling respectively by means of the note container (20), the paper delivery assembly (30), and the printing assembly (40). In this way, when a user desires to operate a service related to a note, the assembled note-issuing apparatus (100) of the present invention enables the operation, thus facilitating the operation, reducing time spent by the user on waiting for a bank teller, and improving working efficiency of bank tellers. In addition, the assembled note-issuing apparatus (100) has advantages of a compact overall structure and small occupied space.
Disclosed in the present invention is a log analysis-based fault location method, the method comprising: reading a trace log, a communication log and a scheduling log of a device to be detected; extracting field data from the trace log, the communication log and the scheduling log; combining the extracted field data having the same time into a log data link list; and detecting the log data link list to locate a fault position. Correspondingly, further disclosed in the present invention is a log analysis based fault location device. The embodiments of the present invention can effectively improve the efficiency of fault location.
A method for monitoring memory data of a drive module. In the method, memory data of a drive module is pre-stored in a memory space with a pre-set structure; and each time a processing request regarding the memory space is received, a current checksum of the memory space is calculated (S101), and an original checksum of the memory space is acquired (S102); if the two checksums are the same, a processing operation can be executed on the memory data (S103); and if the two checksums are different, it indicates that the memory data of the memory space is broken, and a pre-set error processing operation needs to be executed (S104). In the method, memory data of a drive module can be monitored in real time, and an erroneous drive module can also be accurately positioned, thus being able to accurately process a fault in time. Furthermore, a device for monitoring memory data of a drive module is provided, so as to ensure the application and realization of the method in practice.
Provided are a banknote-sorting production line and device thereof for banknote-bundle stamping and code-spraying, comprising a bundle stamping apparatus (10), a bundle code-spraying apparatus (20), and a banknote-supporting platform (220) provided with a banknote inlet (221) and a banknote outlet (222); the bundle stamping apparatus (10) comprises a conveyor belt (110) and a stamping mechanism (120) located above the conveyor belt (110); the bundle code-spraying apparatus (20) comprises a banknote-pushing mechanism (210) and a code sprayer (230) located directly above the banknote-supporting platform (220); the side of the banknote-supporting platform (220) that is adjacent to the banknote inlet (221) is provided with a gap (223) matching the shape of the conveyor belt (110); the outlet end of the conveyor belt (110) extends into the gap (223); the banknote-pushing mechanism (210) comprises a banknote-pushing plate (211) and a first drive mechanism (212) that drives the banknote-pushing plate (211) by moving reciprocatingly from the banknote-inlet (221) side to the banknote-outlet (222) side; the stamping mechanism (120) comprises a stamp (121) and a second drive mechanism (122) that drives the stamp (121) to move. The banknote-bundle stamping and code-spraying device automatically stamps and code-sprays banknote bundles; it has high operating efficiency, each bundle of banknotes has a stamp identifier and a sprayed-code identifier, and in the course of circulation, banknote information can be accurately tracked, making it convenient to refill ATMs.
B65B 61/26 - Dispositifs accessoires, non prévus ailleurs, opérant sur feuilles, flans, bandes, attaches, réceptacles ou paquets pour marquer ou coder les paquets achevés
89.
METHOD AND SYSTEM FOR INTELLIGENT UNINTERRUPTIBLE POWER SUPPLY FOR ATM DEVICE
A method for intelligent uninterruptible power supply (UPS) for an ATM device and a system for intelligent UPS for an ATM device, used for solving the problem of wasted electricity, which is caused by existing UPS systems distributing electricity and power supply time equally and evenly among ATM devices. The method comprises: when a commercial power supply is cut off, acquiring a remaining total electricity of a UPS system (101); acquiring preset priorities of all ATM devices on the UPS system (102); determining the proportion of electricity to be distributed for each ATM device according to the preset priorities of all ATM devices (103); acquiring an available power supply capacity corresponding to each ATM device according to the remaining total electricity and the proportion of the electricity to be distributed (104); estimating a corresponding available power supply time according to the available power supply capacity of each ATM device (105); and controlling the UPS system to power off the ATM device whose power supply time has exceeded said power supply time while the commercial power is cut off (106).
H02J 9/06 - Circuits pour alimentation de puissance de secours ou de réserve, p.ex. pour éclairage de secours dans lesquels le système de distribution est déconnecté de la source normale et connecté à une source de réserve avec commutation automatique
90.
SECURITY ASSOCIATION METHOD AND SYSTEM FOR SELF-SERVICE TERMINAL SECURITY AND PROTECTION SYSTEM
A security association method for a self-service terminal security and protection system. The method comprises: when an ATM system detects a security exception condition, an information interaction system acquiring an alarm signal generated by the ATM system according to the security exception condition (101); the information interaction system sending the alarm signal to a security and protection system so that the security and protection system makes a security and protection response according to the alarm signal (102); when the security and protection system detects that the security exception condition occurs in the surroundings of a self-service terminal, the information interaction system acquiring a security and protection control instruction generated by the security and protection system according to the security exception condition (103); and the information interaction system sending the security and protection control instruction to the ATM system of the self-service terminal, so that the ATM system performs handling as a response according to the security and protection control instruction (104). The method solves the problem that the association advantage of the existing ATM system and security and protection system cannot be used fully due to the fact that the two systems are independent from each other. Further disclosed is a security association system for a self-service terminal security and protection system.
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
G07F 9/02 - Dispositifs pour donner l'alarme ou une indication, p.ex. quand l'appareil est vide; Aménagements publicitaires des appareils déclenchés par des pièces de monnaie
91.
APPLICATION AUTHENTICATION METHOD AND APPARATUS FOR LINUX SYSTEM BASED FINANCIAL SELF-SERVICE DEVICE
Disclosed is an application authentication method for a Linux system based financial self-service device, which is used for solving the problem of a cash security hazard in an existing application authentication method for a Linux system based financial self-service device without an invoking service being added. The method of the embodiments of the present invention comprises: when an upper-layer application program requests to invoke a device interface, acquiring a trusted ID of the upper-layer application program; if the trusted ID of the upper-layer application program is not acquired, prohibiting the upper-layer application program from invoking the device interface; if the trusted ID of the upper-layer application program is acquired, determining whether the trusted ID is in a pre-set trusted list; if so, allowing the upper-layer application program to invoke the device interface; and if not, prohibiting the upper-layer application program from invoking the device interface. Also provided is an application authentication apparatus for a Linux system based financial self-service device, and a financial self-service device.
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
92.
CASH DEPOSIT INLET METHOD AND DEVICE FOR ATM MACHINE HAVING A PLURALITY OF CASHBOXES WITH IDENTICAL DENOMINATION
Provided are a cash deposit inlet method and an ATM machine cash deposit inlet device for an ATM machine having a plurality of cashboxes for identical denominations, which is used to solve the problem that in the existing ATM machines, the cash box located in the front part of the ATM machine is operated frequently, while cash boxes located in the back part thereof are not used. The method comprises the following steps: when the denomination of the plurality of cash boxes in the ATM machine matches the denomination of the banknotes to be put in, the banknotes are put into the cash box having the highest cash-inlet score among the plurality of cash boxes (102), the cash-inlet score of the cash boxes being obtained in advance through the following steps: according to state information of each cash box, scoring each cash box to obtain cash-inlet scores corresponding to each cash box (101), the state information comprising a current count value of a counter of the cash box, and the higher the count value is, the lower the cash-inlet score is.
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
93.
METHOD AND DEVICE FOR CONTROLLING CAMERA AND FINANCIAL EQUIPMENT TERMINAL
The application provides an embodiment of a method for controlling a camera. The embodiment can receive an image transmitted from a camera and perform detection on the image, wherein if a target object is not detected, a search instruction for controlling the camera to rotate in a vertical direction is transmitted to the camera until the target object is detected in an image, and then posture parameters of the target object in the image are acquired, wherein if the posture parameters do not match a preset posture parameter criteria, a fine adjustment instruction for controlling the camera to rotate and move is transmitted to the camera, until the posture parameters of the target object match the preset posture parameter criteria. Thus, the embodiment can perform detection on an image acquired by a camera, i.e., detect whether the image contains a target object and whether a posture of the target object meets an image processing requirement, and if the detection is not successful, adjust the angle of the camera such that the image acquired by the camera meets the requirement. The application also provides a device for controlling a camera and a financial equipment terminal.
Disclosed is a secure communication method for self-service terminal device hardware, which is used for solving the problem that a conventional self-service terminal is vulnerable to a financial crime a criminal commits by simulating device hardware through software to communicate with a host. The method comprises: a local end acquiring original data needing to be sent; the local end encrypting the original data to obtain encrypted data; the local end generating a random number, and partitioning the encrypted data into a data block according to the random number; the local end packaging the data block one by one to generate a corresponding data packet; the local end transmitting the data packet to a peer end, such that the peer end parses and merges the data packet to obtain the encrypted data, and then the peer end decrypting the encrypted data to obtain the original data. Also provided is a secure communication apparatus for self-service terminal device hardware.
G06F 21/85 - Protection des dispositifs de saisie, d’affichage de données ou d’interconnexion dispositifs d’interconnexion, p.ex. les dispositifs connectés à un bus ou les dispositifs en ligne
95.
SELF-SERVICE EQUIPMENT HARDWARE MANAGEMENT METHOD AND DEVICE
An embodiment of the present invention discloses a method for self-service equipment hardware management for solving the inability of the prior art to allow two application programs to coexist further resulting in an inability to accomplish an application scenario requiring simultaneous completion of functions by two application programs. The method thereby alleviates the technical problem of poor control efficiency of equipment hardware control. The embodiment of the present invention comprises: when an application program is running, acquire an equipment and a port required for use by the application program; determine whether the equipment and the port exists, if so, then control the equipment on the basis of the application program and an existing instance, and if not, then make a further determination; the further determination determines whether the port required for use by the application program has been occupied, if so, then control the application program to reassign the port required for use, and if not, then establish a new instance on the basis of the equipment and the port required for use by the application program; and control the equipment on the basis of the application program and the new instance. The embodiment of the present invention also provides a self-service equipment hardware management device.
A method for identifying an identity by means of a social relationship, for use in resolving insufficient security of an existing user account. The method in an embodiment of the present invention comprises: when an identity of a user account needs to be identified, acquiring, by means of a user social relationship database, more than one friend picture of different friend accounts whose social relationships are bound with the user account, the user social relationship database comprising a binding relationship between the user account and the friend accounts; obtaining stranger pictures of stranger accounts whose social relationships are not bound with the user account; disordering the obtained friend pictures and the stranger pictures, and displaying the pictures on an operation interface for a user to select; acquiring the picture selection result of the user on the operation interface; and if the picture selection result is all the obtained friend pictures, passing the identification.
A sheet-type medium gathering device (2) and a financial self-service apparatus. The sheet-type medium gathering device (2) comprises a gathering device housing (101), a gathering device door assembly (106), which can be opened and closed, being connected to the gathering device housing (101) to form a closed space; and a gathering platform (102) located in the closed space, being capable of moving up and down along a guiding mechanism (112); the sheet-type medium gathering device (2) further comprises a pressing plate mechanism, the pressing plate mechanism comprising a flat pressing plate (105). In the process of gathering sheet-type mediums (131), when the gathering device is full of the mediums (131), release, by means of the pressing plate mechanism, the flat pressing plate (105) so as to press the sheet-type mediums (131) on the gathering platform (102) to decrease the thickness of the loose mediums (131), and then continue gathering the mediums (131), thereby increasing the capacity of the gathering device (2).
B65H 29/38 - Délivrance ou progression des articles à la sortie des machines; Progression des articles vers ou dans les piles par bras, châssis, plaques ou éléments analogues mobiles effectuant une opération d'empilage ou d'avancement et avec lesquels le devant des articles est maintenu en contact
B65H 29/68 - Réduction de la vitesse des articles en mouvement
B65H 43/06 - Utilisation de dispositifs de commande de vérification ou de sécurité, p.ex. dispositifs automatiques comportant un élément pour détecter une variable détectant ou réagissant à l'achèvement de la pile
G07D 11/00 - Dispositifs acceptant des pièces de monnaie; Dispositifs acceptant, distribuant, triant ou comptant des papiers de valeur
G07D 13/00 - Manipulation de pièces de monnaie ou de papiers de valeur, caractérisée par une combinaison de mécanismes non couverts par un seul des groupes
98.
SELF-ADAPTIVE IDENTIFICATION METHOD OF IDENTIFYING NEGOTIABLE INSTRUMENT AND DEVICE
A self-adaptive identification method of identifying a negotiable instrument and a device are adopted to resolve problems of a systematic accumulated error and early correction. The identification method comprises: obtaining a sampling parameter, and sampling, according to the sampling parameter, an optoelectronic signal of a negotiable instrument (101); obtaining an optoelectronic signal correction value, and performing, according to the optoelectronic signal correction value, digital compensation on the optoelectronic signal (102); extracting a characteristic from the digitally compensated optoelectronic signal to obtain an eigenvector (103); inputting the eigenvector into a predefined classifier to perform identification, so as to obtain an identification result of the negotiable instrument (104); obtaining, according to the identification result, a specific area corresponding to the negotiable instrument (105); obtaining, according to the specific area, characteristic information of the optoelectronic signal of the negotiable instrument (106); computing, according to the characteristic information, an accumulated component and a differentiation error of the negotiable instrument (107); computing, according to the accumulated component and the differentiation error, a total correction value of the optoelectronic signal (108); updating, according to the total correction value, the correction value and the sampling parameter of the optoelectronic signal (109); and outputting the identification result (110).
A sheet medium flattening device (60) and financial self-service device. The sheet medium flattening device comprises: an installation frame (1); an annular channel (2), comprising an annular inner channel plate (21) and an annular outer channel plate (22), wherein the inner channel plate (21) has a hollow cavity provided with a heating module (23) therein, the heating module (23) provides heat to the inner channel plate (21), the outer channel plate (22) is an annular guiding plate surrounding the inner channel plate (21), and the outer channel plate (22) is provided with a sheet medium opening (221) abuttingly joined to a delivery channel (62) for the sheet medium; at least one pinch roller (3), forming a pinch with the inner channel plate (21); a reversing mechanism (4), disposed at the sheet medium opening (221) for realizing selectively delivering of paper money between the delivery channel (62) and the annular channel (2) of the flattening device (60); a driving device (5) and a control system. The flattening device (60) can reduce the level of crumpling of the paper money effectively and therefore improve the reliability and stability of processing that a financial self-service device performs on the paper money such as delivery, identification, and stacking.
G07D 13/00 - Manipulation de pièces de monnaie ou de papiers de valeur, caractérisée par une combinaison de mécanismes non couverts par un seul des groupes
100.
SELF-SERVICE EQUIPMENT ENERGY SAVING CONTROL METHOD AND DEVICE
An energy saving control method and a device for a self-service equipment, are used for solving issue of the sleep interval being too short and the self-service equipment repeatedly going to sleep and restart when the amount of users is large, and the sleep interval being too long when the amount of users is small, resulting in waste of resources. The method comprising: obtaining to-be-learned sample usage information from historical data of the self-service equipment users (101); learning the to-be-learned sample information according to a preset Bayesian Prior Probability Model to obtain a learning result (102); updating the Bayesian Prior Probability Model on the basis of the learning result (103); predicting the amount of users for each time period in a preset time period by means of using the updated Bayesian Prior Probability Model to obtain a predicted amount of users of the self-service equipment (104); and adjusting a sleep interval for each time period of the self-service equipment in accordance with the predicted amount of users (105).