79 results about "Certificate" patented technology

The invention discloses an efficient certificateless signature method based on an elliptic curve discrete logarithm problem. The method has the advantages of a common certificateless signature algorithm, namely a certificate management problem of a traditional public key system and a key escrow problem in an identity password system are eliminated; and the algorithm is simple in calculation process, and time-consuming bilinear operation does not need to use. The algorithm can be efficiently applied to fields such as e-commerce and the like of requiring the electronic signature.

The invention relates to an artificial intelligence technology, and discloses a certificate recognition method based on deep learning, which comprises the following steps: detecting a to-be-recognized image by using a detection model based on a YOLO algorithm to obtain a certificate block diagram; intercepting the certificate frame diagram to obtain a certificate diagram; performing semantic segmentation on the certificate image by using a pre-trained certificate semantic segmentation model to obtain a plurality of key field boxes; and performing character recognition on the certificate image according to the key field box to obtain certificate information. In addition, the invention also relates to a blockchain technology, and the to-be-recognized image can be stored in a node of a blockchain. The invention further provides a certificate recognition device based on deep learning, electronic equipment and a computer readable storage medium. According to the invention, the problem of low certificate identification accuracy can be solved.

The embodiment of the invention provides a method and system for generating an anonymous certificate, and the method comprises the steps: generating the root hash of a Merkel tree according to the identity information of a user; signing the root hash, and sending the signed root hash to a request end, so that the request end compares the signed root hash with a root hash generated by a preset algorithm, and sends an acknowledgement signal if the signed root hash is the same as the root hash generated by the preset algorithm; and if the acknowledgement signal is received, generating an identitycertificate of the user by utilizing the Merkel tree. According to the method and the system for generating the anonymous certificate, the identity certificate is generated through the Merkel tree, so that only the hash value of the identity information is stored in the identity certificate, and leakage of private data of a user is avoided to a certain extent.

The embodiment of the invention provides a certificate classification method, device and equipment. According to the scheme, the method comprises the following steps of: carrying out preliminary identification on an obtained certificate image by adopting a classification model to obtain a plurality of confidence scores for various certificate types, performing secondary comparison on the pluralityof candidate certificate types of which the confidence scores are greater than a first preset score and the to-be-identified certificate, specifically, respectively extracting a plurality of featurevectors of a mark region and a fixed character region in the plurality of candidate certificate types, and then extracting feature vectors of the mark region and the fixed character region in the to-be-identified certificate, and calculating the similarity between the first feature vector and the plurality of second feature vectors, and determining the candidate certificate type corresponding to the maximum similarity value as the type of the to-be-identified certificate in the plurality of obtained similarity values.

The invention discloses a certificateless ring signcryption method based on a multiplication group. The certificateless ring signcryption method comprises the steps of initializing a system, generating a private key and a public key of a user, generating part of the private key of the user, ring signcryption and de-signcryption. An encryption and anonymous authentication method based on a multiplication group is adopted in the ring signcryption step, so that the complex certificate management problem in the traditional public key infrastructure, the key escrow problem in identity cryptographyand the anonymous communication problem in a certificateless mode are solved; based on certificateless public key cryptography and a multiplication cycle group, the safety is improved, the problems ofdependence on a secure channel, certificateless management and key escrow are avoided, adaptive selection ciphertext attacks and adaptive selection message attacks can be resisted, and the method issuitable for a network information safety application environment. The method has the advantages of being good in safety, low in calculation complexity and the like, and can be applied to the field ofnetwork information safety.

The invention discloses a certificate information detection and extraction method. A synthesized certificate data set is used to replace a real certificate data set for the training of a Faster-RCNN detection model; certificate information is detected and classified by using a method based on a deep neural network, and special preprocessing is not needed; and an end-to-end character recognition method is used for recognizing the certificate character information with the uncertain length, and segmentation is not needed. The character length does not need to be set, a single Chinese character does not need to be recognized, the influence on the recognition rate due to segmentation errors is avoided, and the text information can be obtained only by inputting a picture into the network model.Compared with a traditional method based on character template matching, the method has strong adaptability when facing the problems of low brightness, low contrast, uneven illumination, deformation,incompleteness, shielding and the like, and the precision of the method is far superior to that of the traditional method.

The invention relates to an accounting document intelligent management method which comprises the following steps: establishing a category sample library and a position parameter library, wherein thecategory sample library comprises a corresponding relationship between feature values and original certificate categories, and the position parameter library comprises position parameters of all key fields in different types of original certificates; obtaining an original certificate picture, wherein the original certificate picture comprises a plurality of key fields; extracting an original certificate characteristic value; querying the type of the original certificate in a type sample library according to the characteristic value; querying position parameters of each key field in a positionparameter library according to the type of the original certificate; constructing and training a deep learning model; positioning and extracting an area where each key field is located according to the position parameters, and identifying each key field by using the trained deep learning model; searching an accounting voucher template and a reimbursement voucher template; filling the accounting voucher template and the reimbursement voucher template with the key fields to generate an accounting voucher and a reimbursement voucher.

The invention relates to a gambling domain name recognition method based on certificates and domain name resolution, which solves the technical problem that the existing gambling domain name recognition accuracy is low under the condition that a webpage text is not resolved, and comprises the following steps of: establishing a Chinese classification model constructed based on Bert fine tuning; respectively carrying out digital certificate analysis on the Chinese gambling domain name and the benign domain name; obtaining a text feature vector of the domain name through an N-gram method; and training and testing the digital certificate analysis feature vector and the domain name text feature vector of the Chinese gambling domain name and the benign domain name through RNN (Recurrent Neural Network), Decision Tree, ExtraTree, RandomForest, KNN (Kalman Neural Network) and SVM (Support Vector Machine) learning algorithms, and constructing a Chinese gambling domain name mining model. The method can be widely applied to identification of Chinese gambling domain names.

The invention relates to the technical field of data processing, and the certificate making efficiency is improved through multi-level caching. The invention specifically discloses a voucher generation method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a voucher rule from a database, and respectively storing the voucher rule in a preset first-level cache or second-level cache according to the rule type of the voucher rule; if the to-be-certificated data is received, determining a service type identifier corresponding to the to-be-certificated data; reading voucher rules corresponding to the service type identifiers from the second-level cache, and judging whether the voucher rules corresponding to all the service type identifiers are read or not; if the voucher rule corresponding to the service type identifier is not read, reading the voucher rule corresponding to the service type identifier from the first-level cache according tothe service type identifier which is not read; and performing certificate making on the to-be-made certificate data according to the certificate rules corresponding to all the business type identifiers to obtain a certificate corresponding to the to-be-made certificate data.

The invention discloses an academic achievement pre-publishing and right authentication system and method based on a block chain, and belongs to the technical field of block chain application. The problem of how to carry out right confirmation and authorization work on academic achievements by using the block chain is solved. The method is characterized by enabling a user to register in a real-name system; after logging in, uploading a work to be chained; compressing works through a feature extraction algorithm, converting the works into character strings for Hash operation, transmitting a Hash value to a block chain, storing original works in a database, checking duplication in an uploading process, and checking whether the works are stored by an upper chain, are determined not to be protected or are similar to existing works in the database, and preventing plagiarism infringement behaviors; if the works do not pass the system auditing, refusing to link, and returning the refused reason to the user; and chaining the works automatically through duplicate checking and auditing, and issuing an electronic certificate to the user. According to the method and the system, the problem of publishing delay is avoided and the right attribution problem of academic achievements is reduced by utilizing the characteristic of the block chain based on Internet access.

The invention relates to a lot-drawing method for electing a consensus node. The method comprises the following steps of: respectively signing a seed parameter by using a private key of each node, andprocessing a signature result according to a random algorithm to obtain a first parameter; performing fusion processing on the PoW, the PoS and the PoC of each node to obtain a second parameter; according to the first parameter and the second parameter corresponding to each node, obtaining a certificate of each node; and according to the comparison result of the certificate of each node and the entry probability, obtaining the consensus node. According to the method, a generalized workload proving method is introduced into the lot-drawing algorithm for selecting the consensus participation node, and a lot-drawing mechanism is reinforced and supplemented for various attack modes, and the drawing mechanism has been strengthened and supplemented for a variety of attack methods, so that potential safety and stability hazards of the lot-drawing algorithm when the lot-drawing algorithm is subjected to network attack and market fluctuations are eliminated.

The invention belongs to the technical field of certificate identification, and particularly relates to a multi-template certificate rapid certification method and system, a medium and a device. The invention aims to solve the problems of long customs quarantine auditing time and low accuracy in the prior art. The method comprises the following steps: receiving a certificate picture to be audited; matching a template corresponding to the certificate picture; based on coordinates of a preset ROI region in the template, frame-selecting an actual ROI region in the certificate picture; processing the frame-selected picture of the actual ROI region through an OCR model to obtain a field result in the actual ROI region; extracting the field result to obtain final certificate information; and judging the authenticity of the certificate to be audited according to the identified certificate information. According to the invention, the actual ROI of the picture is obtained based on the textbox matching template and the template preset ROI, and then the actual ROI is identified, so that the interference of other miscellaneous items is avoided, and the identification accuracy is improved.

The invention relates to an image processing technology, and discloses a certificate positioning method, device and equipment and a storage medium, and the method comprises the steps of marking the vertex coordinates of a certificate frame in a certificate sample image, and respectively generating a first thermodynamic diagram, a first scale mapping diagram and a first center offset diagram; performing prediction processing on the certificate sample image by using a pre-constructed certificate positioning model, and respectively obtaining a second thermodynamic diagram, a second scale mapping diagram and a second center offset diagram which are obtained by prediction; calculating a first loss value, a second loss value and a third loss value; calculating a loss sum according to a loss formula; performing repeated iteration training on the certificate positioning model, recalculating the loss sum, and when the loss sum is smaller than a preset threshold value, obtaining a standard certificate positioning model; and positioning the to-be-positioned certificate image by using the standard certificate positioning model to obtain a certificate frame in the to-be-positioned certificate image. According to the invention, the certificate frame can be accurately positioned from the certificate image with a complex background.

The invention discloses a certificate classification method, which comprises the following steps: when the placement operation of a target certificate is detected, according to the certificate size ofthe target certificate, determining one or more certificate identifier extraction positions; extracting a certificate identifier based on the certificate identifier extraction position, and confirming a valid certificate identifier; and determining the certificate type according to the certificate identifier extraction position corresponding to the effective certificate identifier. The inventionfurther discloses a certificate classification device and a computer readable storage medium. Through a new certificate classification method, a certificate classification step is set to determine a current certificate type; related certificate type information can be added/modified based on certificate parameters of certificates in application; a large amount of numerical support is not needed; the certificate classification method is low in maintenance cost and small in occupied CPU during running; no specified running equipment is needed; popularization is facilitated; and the beneficial effect of improving certificate recognition efficiency is achieved.

According to the invention, a 1+X certificate alliance chain, a cryptographic operation module, a federal learning module, an applet, a storage module, an image processing input module, an image processing output module, a sound processing input module, a sound processing output module, a fingerprint identification module, a employment demand and talent matching degree algorithm and an intelligent contract code are established based on WeBank FISCO-BCOS. An asymmetric key pair is generated through the cryptography operation module, a public key, a coded lock number, learning and practical training record types, purposes, ranges and owner information are registered in a block chain, assets are bound with identity information, dual authentication is performed through CA, a block chain technology and a federated learning technology are used, the occupational levels of talents are fairly evaluated through a machine learning algorithm, a federated machine learning model is stored on a federated learning block chain, and the whole process of data and the machine learning model is recorded and traced. The invention has the beneficial effects that the evaluation is accurate, and cheating is avoided.

The invention is applicable to the technical field of digital certificate security and provides a certificate issuing system based on an SM2 algorithm. The certificate issuing system based on the SM2 algorithm includes a certificate application service platform used for providing network security guarantees; and an SM2 trust service platform which is used for providing an authentication service to the certificate issuing system. In the certificate issuing system, a certificate release platform is designed to ensure strict separation of a service application service and certificate release so that it is convenient for statistics of certificate information. A Quartz technical framework which is higher in efficiency, lower in coupling and better in transportability is adopted to realize CRL regularly released tasks and thus development time is saved and a large quantity of Jqery and ajax technologies are adopted to ensure that in a registration submission process, data format check and data uniqueness check are both carried out on a page to improve significantly the interface friendliness.

The invention provides a data tamper-proofing method based on an abstract algorithm, and the method comprises the steps: 1, extracting fields, needing to participate in the generation of a digital certificate, in a data table, and carrying out the character string splicing of the fields, and obtaining an original character string; 2, setting a secret key md5Key, and splicing the original characterstring and the secret key md5Key to form a new character string; 3, enabling the new character string to be subjected to an MD5 digest algorithm, and generating a digital certificate, wherein the digital certificate is a number within 11 bits; 4, generating a digital certificate, and storing the digital certificate into a digital certificate field of the current data table; 5, when the delivery operation is carried out, repeatedly executing the steps 1 to 3, comparing the newly generated digital certificate with the digital certificate field stored in the data table, if the newly generated digital certificate is consistent with the digital certificate field stored in the data table, legitimate, allowing the delivery, and if the newly generated digital certificate is inconsistent with thedigital certificate field, not legitimate and not allowing the operation. Safety guarantee is provided for the system.

The invention discloses a certificate information acquisition method and device, a storage medium and computer equipment, relates to the technical field of image recognition, and mainly aims to solve the problem of low certificate information acquisition efficiency in the prior art. The method comprises the steps: acquiring certificate image information, and identifying text content information and textbox position information in the certificate image information; and performing quantization processing on the text content information and the textbox position information, and performing classification processing on the text content information and the textbox position information after quantization processing by using a trained classification model, enabling the classification model to update text content information or textbox position information in a fuzzy training sample set according to different fuzzy replacement coefficients in the training process; and storing the text content information subjected to classification marking into a certificate data list with different certificate characteristics according to a classification processing result.

The invention discloses a character positioning method and system for a certificate document, and the method comprises the steps of inputting a to-be-detected image to a deep learning model, and outputting a spliced and fused feature map; performing image difference operation on the feature maps marked by different indexes to obtain a feature map difference; carrying out binarization calculation on the difference value of the feature map; performing connected domain marking on the image according to a binarization calculation result; and traversing all connected domains, converting the connected domains into positioning coordinates, and outputting a coordinate set. According to the invention, automatic identification of license document characters is realized, the positioning precision ishigh, and character detection in any direction is supported.

The invention provides an electronic certificate retrieval method and device, and a computer readable medium. The method comprises the steps: obtaining basic information and service information of each first electronic certificate; determining identification information of each piece of service information; creating a corresponding ID for each first electronic certificate; generating an electroniccertificate data model; randomly extracting at least one second electronic certificate from each first electronic certificate, and obtaining service information and identification information of eachsecond electronic certificate; determining a first index model; processing the service information and the identification information of each second electronic certificate in the first index model according to a decision tree algorithm to obtain a decision tree; inputting a retrieval scene of the to-be-retrieved electronic certificate data into the decision tree to determine a second index model;and determining a target ID corresponding to the retrieval scene by utilizing a second index model so as to obtain the to-be-retrieved electronic certificate data from the electronic certificate datamodel. According to the scheme, quick retrieval of the electronic certificate can be realized.

The invention relates to the technical field of artificial intelligence, and provides a missed number detection method and device, electronic equipment and a storage medium. The method includes: determining the execution state according to the state identifier of the current task; when the triggering condition is met, timely performing missing number detection; when the current task is executed completely, obtaining all certificate numbers in the current task, avoiding erroneous detection, determining the maximum voucher number and the total number of all voucher numbers; when the maximum certificate number is not equal to the total number, assigning the maximum certificate number to a preset variable as a variable value; utilizing a missing number detection model constructed through a reverse insertion method to continuously adjust variable values based on all the certificate numbers, and obtaining missing numbers. Through automatic data processing, the missed number detection accuracy is improved, the missed numbers are recorded into a missing number table, and automatic detection of the missed numbers is achieved based on artificial intelligence. The invention also relates to ablockchain technology. The missed number detection model can be deployed in the blockchain.

The invention relates to the technical field of artificial intelligence detection models, and provides a federal model training method and device, a certificate detection method and device, equipment and a medium, and the method comprises the steps: obtaining a participant complete set initial certificate federal model; screening out a qualified participant subset from the participant complete set according to a preset screening scheme; sending the initial certificate federation model to a participating terminal; and receiving returned model parameters and performance consumption parameters. The Fedavg algorithm is used for aggregating the model parameters to obtain model iteration parameters, and meanwhile the contribution degree and the excitation value are determined according to the model iteration parameters, the model parameters and the performance consumption parameters through the contribution excitation model; and when a preset convergence condition is not met through detection, the qualified participant subset is updated, and the initial certificate federated model is iteratively updated according to the model iteration parameters until convergence is performed to obtain a certificate detection model. According to the invention, the terminal is excited to participate in model training through the contribution degree and the excitation value, and the efficiency and accuracy of federal learning are improved.

The embodiment of the invention provides a test sample generation method, and the method comprises the steps: configuring at least one target certificate base map for a target certificate category, and enabling the target certificate base map to be a certificate picture which does not carry feature data; obtaining a plurality of target features associated with the target certificate category; obtaining a plurality of target feature data sets from a database according to the plurality of target features, wherein each target feature data set comprises a plurality of target feature data corresponding to the plurality of target features; and generating a plurality of test samples according to the target certificate base map and the plurality of target feature data sets, and each test sample comprises the target certificate base map and a plurality of target feature data corresponding to one target feature data set in the plurality of target feature data sets loaded on the target certificate base map. According to the test sample generation method provided by the embodiment of the invention, a large batch of certificate pictures which meet format requirements and do not relate to privacy in multiple scenes can be automatically generated through a small amount of data.

The invention relates to the field of artificial intelligence, discloses a certificate picture generation method, device and equipment and a storage medium, which are used for improving the accuracy of generating a certificate picture conforming to a real scene. The certificate picture generation method comprises the steps of obtaining a sample certificate picture, wherein the sample certificate picture comprises sample text data and sample background data; using a picture similarity comparison algorithm for forming certificate background data and certificate character data based on the sample certificate picture, and the certificate character data comprising character language data and font style data; writing the certificate text data into the random position of the certificate background data to generate an initial certificate picture; preprocessing the initial certificate picture to generate a plurality of preprocessed certificate pictures; and adopting a preset random scaling function to perform multiple times of random scaling on the plurality of preprocessed certificate pictures, so that a plurality of target certificate picture groups can be generated. In addition, the invention also relates to a blockchain technology, and the sample certificate picture can be stored in the blockchain.

The invention relates to the technical field of artificial intelligence, in particular to a multi-certificate segmentation association method based on intelligent decision and related equipment. The method comprises the following steps: acquiring a pre-segmented picture for segmentation, and inputting the pre-segmented picture into a picture segmentation model to obtain a plurality of certificatesub-pictures and certificate types; respectively carrying out character recognition processing on the single certificate subgraph to obtain a plurality of feature data in the certificate subgraph; andassociating the feature data in the plurality of certificate subgraphs to obtain associated information among the plurality of certificate types. Certificate pictures are automatically processed andsegmented, different types of certificates are automatically classified, different certificate types are obtained, and single certificate can be independently used. Through organic association of different certificate types, data is provided for subsequent retrieval of user information.