209results about "Signature reading/verifying" patented technology

The automatic tax declaration and recheck system comprises a character recognition module, a process automation module, a dynamic rule engine module, a block chain evidence storage module and a distributed database. The character recognition module forms a tax field signal, a structured signal and a semantic tag signal. And the process automation module forms a tax declaration form according to the tax field signal. And the dynamic rule engine module receives the structured signal, performs multi-dimensional verification through a machine learning model, and forms a verification signal. The block chain evidence storage module comprises a distributed database, and the block chain evidence storage module receives the semantic tag signal and forms a unique hash value. And the distributed database performs feature fusion on the unique hash value, the tax declaration form and the verification signal to form a declaration signal and stores the declaration signal in the distributed database. According to the automatic tax declaration and rechecking system, the problem that structural analysis of multi-format tax documents is difficult can be solved.

A system and a method for verifying authenticity of a physical copy and a digital copy of a document are disclosed. The method comprises registering a document in a repository by storing details related to the document in a location of the repository. A symbology for the document is generated. The symbology is an identifier of the location of the repository comprising the document. The symbology is associated with either a physical or a digital copy of the document. The digital copy of the document is printed to generate a printed copy. The printed copy or the physical copy of the document is scanned to generate a scanned image. The document and the details related to the document present at the location of the repository are accessed. The scanned image is compared with the document stored in the repository to determine the authenticity of either the physical copy or the digital copy of the document.

The invention relates to the technical field of information security, and discloses a handwriting dynamic encryption and electronic signature method based on a national secret algorithm, and the method comprises the steps: obtaining the handwriting data of a user, and carrying out the preprocessing of the data to remove noise and errors; carrying out feature extraction on the preprocessed handwriting data, extracting corresponding track features, speed features and pressure features, and generating a feature hash value of a unique identifier through a cryptographic SM3 hash algorithm; splicing and combining the feature hash value and a user private key, deriving a dynamic key through a national secret SM2 algorithm, and generating an electronic signature by using the dynamic key; an electronic signature is combined with a timestamp to form a structured signature packet, and the signature packet is hierarchically encrypted and stored in a database through a national cryptographic SM4 algorithm. According to the method and the system, the handwritten signature picture, the digital signature technology, the electronic signature technology and the biological recognition and layout file processing technology are fused, so that the safety and the reliability of electronic document signing are improved.

The invention relates to the technical field of data management, in particular to an insurance document data management method and system and an intelligent terminal. The method comprises the following steps: acquiring insurance document data uploaded by a user, and marking the insurance document data to obtain checked insurance document data; constructing an insurance document labeling agent according to the checked insurance document data; acquiring insurance document data uploaded by a user in real time, and classifying the insurance document data to obtain normally-labeled document data and abnormally-labeled document data; evaluating the abnormal labeled certificate data to obtain high-deviation labeled certificate data and low-deviation labeled certificate data; reconstructing the high-deviation labeled certificate data to obtain credible high-deviation labeled certificate data; correcting the low-deviation labeled certificate data and the credible high-deviation labeled certificate data to obtain deviation correction labeled certificate data; and packaging the normal labeling certificate data and the deviation correction labeling certificate data. According to the invention, the efficiency, accuracy and automation level of insurance document management are improved.

The present disclosure provides a character recognition model training method and apparatus, a character recognition method and apparatus, a device and a medium, relating to the technical field of artificial intelligence, and specifically to the technical fields of deep learning, image processing and computer vision, which can be applied to scenarios such as character detection and recognition technology. The specific implementing solution is: partitioning an untagged training sample into at least two sub-sample images; dividing the at least two sub-sample images into a first training set and a second training set; where the first training set includes a first sub-sample image with a visible attribute, and the second training set includes a second sub-sample image with an invisible attribute; performing self-supervised training on a to-be-trained encoder by taking the second training set as a tag of the first training set, to obtain a target encoder.

Aspects of the embodiments described herein are related to systems, methods, and computer products for performing automatic check fraud detection. Aspects of embodiments described herein provide artificial intelligence systems and methods that analyze an image of an incoming check, compare the features of the image to the associated features on a reference check, and generate a check fraud score based on the comparison, and either approve the incoming check or flag the incoming check for manual review.

The invention discloses a calligraphy practice auxiliary system and method based on an intelligent visual system, and the method comprises the steps: firstly, synchronously collecting the three-dimensional track, wielding angle and pressure distribution data of a pen point through a high-speed camera and an inertial measurement unit, comprehensively recording the actions of a real pen and a virtual pen, and solving the defect that the traditional system is difficult to capture the virtual pen in the air; secondly, key indexes such as potential energy density and rhythm entropy are quantified through space-time sequence modeling, and dynamic coherence and rhythm changes of cursive scripts are accurately analyzed; and thirdly, an ideal virtual pen path is generated based on the light-weight GAN network, and the deviation degree is calculated, so that a more intuitive improvement basis is provided for a user. And finally, aligning the speed-pressure distribution matrixes of the user and the famous family by means of a dynamic time warping algorithm, quantitatively evaluating the charm coherence, projecting a three-dimensional track and a charm thermodynamic diagram in real time, and providing personalized training suggestions. According to the method, the limitation that a traditional calligraphy auxiliary technology depends on static trajectory analysis is broken through, and refined capture and quantitative evaluation of cursive writing dynamic writing are achieved.

The invention provides a power plant work ticket static inspection method and system. The method comprises the following steps: scanning a work ticket to be inspected to obtain a work ticket image, identifying the work ticket image, establishing a work ticket list, and screening a work ticket template corresponding to the work ticket list; checking whether the to-be-signed position is signed according to the work ticket template, and if yes, determining the authenticity of the signed handwriting; if the handwriting is true, checking whether the time to be filled is filled according to the work ticket template, and if so, determining whether the steps executed according to the work ticket are continuous in the forward direction; if the positive direction is continuous, determining whether the safety measures in the work ticket list are correct and complete, and whether the safety measures are executed; if the safety measures are correct and complete and are executed, whether the stamping position and content are correct or not is recognized; and if the stamping position and content are correct, the to-be-inspected work ticket is determined to be qualified, and the inspection accuracy and the inspection efficiency of the work ticket are effectively improved by automatically scanning, identifying and verifying the to-be-inspected work ticket.

The invention provides a title review material auditing method and device, a terminal and a storage medium, and belongs to the technical field of auditing authentication, and the method comprises the steps: receiving a plurality of pieces of review related information and corresponding title review materials; generating a first prompt word according to all the review related information; preprocessing all the title review materials to obtain a plurality of pictures, and inputting all the pictures and a first cue word into a multi-modal large model to obtain a first judgment conclusion; after extracting texts from all the pictures, inputting the texts and the first cue word into the semantic large model to obtain a second judgment conclusion; and based on the first judgment conclusion and the second judgment conclusion, determining a final conclusion about the correlation between the review related information and the title review material, and outputting the final conclusion. According to the invention, the correlation between the review related information and the title review material is judged by using the two models, and the correlation between the review related information and the title review material is determined based on the output results of the two models, so that the review efficiency can be effectively improved.

Systems, apparatuses, methods, and computer program products are disclosed for providing continuous signature management. An example method includes detecting, by an event identification circuitry, a user authentication event and receiving, by communications hardware, a signature from the user during the user authentication event. The example method further includes determining, by the event identification circuitry, a signature parameter set for the signature and determining, by the event identification circuitry, whether the user authentication event corresponds to an authentic event type. The example method further includes in an instance in which the user authentication event corresponds to the authentic event type, storing, by signature management circuitry, the signature and corresponding signature parameter set in a signature profile associated with the user within a signature corpus.

Disclosed herein are computer-implemented method, system, and computer-readable medium embodiments for providing a virtual second image during remote deposit of documents. A system generates, from a field of view of a camera, a live stream of image data comprising imagery of a portion of the document. The system obtains, via an optical character recognition program, one or more data elements from the imagery of the portion of the document and generates a first image of a front side of the document. The system receives an electronic signature of a customer via a user interface and, upon verifying the electronic signature, generates a virtual second image by merging the electronic signature with a second image for a back side of the document selected from a plurality of back images. The system transmits the first image, the virtual second image, and the one or more data elements to a remote deposit platform.

The invention relates to the technical field of risk assessment, in particular to a bill risk assessment method and system for a bill comprehensive management platform based on artificial intelligence, and the method comprises the following steps: positioning and recognizing bill information based on image edge features and field anchor points, verifying identity consistency, constructing a role path chain, and verifying a circulation logic relation. And marking behavior anomalies and paths. According to the method, stable structure reference points are extracted through image edge ink mark concentration changes, tilt correction and size adjustment are carried out, accurate positioning of a field area is achieved in combination with a field anchor point proportion, structured information extraction and field content recognition are achieved, and character consistency judgment and seal superposition degree matching are combined to execute identity verification. Through verification of a node sequence and a role evolution path logic relation, identification of abnormal behaviors in a transaction path is realized, identity consistency is verified, transaction chain rationality is evaluated, and risk identification granularity and path behavior abnormity determination precision are improved.

The embodiment of the invention discloses a handwriting image identification method and system based on artificial intelligence, and the method comprises the steps: firstly collecting a to-be-identified handwriting image and a known sample handwriting image, and generating a target handwriting image set; sequentially performing stroke region segmentation, stroke feature extraction and texture mode analysis on the handwriting images in the set to generate a mixed feature data set containing a traditional handwriting feature set and a deep learning handwriting feature set; inputting the mixed feature data set into a Triplet network model, and generating a first discriminative feature vector of the to-be-identified handwriting image and a second discriminative feature vector of a known sample handwriting image through triple loss function joint optimization training; and finally, generating a handwriting image identification report containing the feature matching score and the author affiliation probability based on the first discriminative feature vector and the second discriminative feature vector. The accuracy and reliability of handwriting identification are improved.

The embodiment of the invention discloses a user verification method and device, a program product and a storage medium, relates to the field of artificial intelligence, is also applicable to the field of financial science and technology, and comprises the following steps: obtaining handwritten image data of a to-be-verified user; the handwritten image data comprises a to-be-verified user identifier, handwriting platform information, handwriting process parameters and a handwritten image; performing content verification on the handwritten image based on the to-be-verified user identifier to obtain a content verification result of the handwritten image; the content verification result is passing or not passing; when the content verification result is successful, determining a target handwriting verification model based on the to-be-verified user identifier, and inputting the handwritten image data into the target handwriting verification model to obtain verification data output by the target handwriting verification model; and determining a handwriting verification result based on the verification data, and displaying the handwriting verification result to the to-be-verified user. According to the invention, the risk that the biological characteristic information of the user is leaked can be reduced, and the safety and accuracy of user identity verification are improved.

To perform learning quickly with a small amount of training data and to verify signatures with high accuracy.SOLUTION: A signature verification system is as described in the following. An image containing one or more handwritten characters is divided into a plurality of patches of a predetermined size for each character. For each of the divided patches, an embedding feature, which is data representing pixel and positional features of the patch, is calculated. Based on the calculated embedding features of each patch, the type of each patch is identified, and based on the identified patch types and the positions of the patches, a context of each patch in the image is calculated. A discrete representation model that applies an attention mechanism, MSA, to the overall embedding features is stored. A signature verification model is provided, which receives, as input, two images each containing one or more handwritten characters and outputs a probability that the two input images are images of characters written by the same writer. A new signature verification model is generated by replacing a part of the signature verification model with the discrete representation model.SELECTED DRAWING: Figure 4

The invention provides a multi-model signature matching method and device, a storage medium and a program product, and relates to the technical field of text recognition. The multi-model signature matching method comprises the following steps: preprocessing a signature image; extracting a feature vector of the signature image; performing OCR model and multi-modal large model identification on the signature image; performing similarity value calculation on the vector of the signature image and the stock vector; and the similarity values between the vectors and the recognition results of the OCR model and the multi-modal large model are integrated to be matched, and a final result is obtained. According to the method, the feature extraction model, the OCR model and the multi-modal large model are used, and the results of various models are combined according to a certain algorithm, so that the matching precision of signature images such as poor signature images and fuzzy signature images is effectively improved.

The invention relates to a visual question and answer method and device, electronic equipment, a storage medium and a computer program product. The method comprises the steps of obtaining a target image and a target question for the target image; generating an initial program based on the target problem; adding a target code for recording a program execution process to the initial program; inputting the target image into an extension program, and obtaining a prediction answer for the target question, execution process information of the extension program and a screenshot image; based on the execution process information and the screenshot image, an interpretation in a multi-modal form for the predicted answer is generated. Therefore, the decision basis picture and the semantic association analysis can be synchronously generated while the prediction answer is output, that is, the explanation of the reasoning process of the prediction answer can be increased and output, so that a user can intuitively understand the corresponding relationship between the image features and semantic reasoning, the reasoning transparency and the decision credibility can be improved, and the user experience can be improved. Therefore, popularization and application of the visual question-answering technology are facilitated.

The invention discloses a signature handwriting authenticity digital identification method based on multi-modal features, and relates to the technical field of image processing, and the method comprises the following specific steps: 1, obtaining off-line image data and on-line time sequence data of two groups of signatures to be identified, the data preprocessing module carries out standardization processing on the offline image data and the online time sequence data of the signature to be identified to obtain preprocessed offline image data and preprocessed online time sequence data; 2, inputting the preprocessed offline image data into an offline feature extraction network, and extracting and outputting 1024-dimensional offline features; and 3, inputting the preprocessed online time sequence data into an online signature feature extraction network. According to the method, the online features and the offline features of the signature handwriting are fused and combined to form multi-mode signature handwriting digital recognition, the subtle difference of the counterfeited signature handwriting in static structure consistency and dynamic behavior consistency can be effectively captured, and the recognition capability of complex counterfeited means is remarkably improved.

The invention provides a character recognition management method and system fusing multi-modal data, and relates to the technical field of data processing, and the method comprises the steps: continuously capturing heterogeneous multi-source unsteady business instruction sequences submitted by employees; based on the heterogeneous multi-source unsteady service instruction sequence, executing a tensor reconstruction operation of an embedded space through a cross-modal depth feature alignment network, and outputting a structured high-dimensional tensor of a dimension specification; and based on the structured high-dimensional tensor, performing homogenized set division by using a K-means clustering algorithm driven by feature similarity to generate a feature set cluster with a compact feature. According to the invention, efficient management of text information in a banking business scene is realized.

Systems and methods are provided for processing and extracting content from an image captured using a mobile device. In one embodiment, an image is captured by a mobile device and corrected to improve the quality of the image. The corrected image is then further processed by adjusting the image, identifying the format and layout of the document, binarizing the image and extracting the content using optical character recognition (OCR). Multiple methods of image adjusting may be implemented to accurately assess features of the document, and a secondary layout identification process may be performed to ensure that the content being extracted is properly classified.

The invention belongs to the field of image recognition and processing, particularly relates to a handwritten signature set construction method and system based on a protocol and computer equipment, and aims to solve the problem of how to improve the construction efficiency of a handwritten signature data set. The method comprises the following steps: performing text content identification on a protocol image file to obtain a text region block, a position of the text region block, and characters and character positions in the text region block; extracting characters in the text region blocks and splicing the characters to obtain text content; identifying a printed name and a handwritten signature in the text content, and matching the printed name with the handwritten signature to obtain a name pair set; and traversing and verifying the name pair set to obtain a labeling result of the handwritten signature. According to the method, the whole process automation of handwritten signature recognition, verification and labeling is realized, and the efficiency and accuracy of handwritten signature set construction are improved.

The application discloses a kind of doctoral and master degree thesis authorization page identification method, comprising: extracting electronic version degree thesis by page to form image set to be examined;According to the image set to be examined, paper version degree thesis is scanned to generate image set to be examined;Image set to be examined is sent into authorization page identification classifier, it is judged whether there is authorization page;If not, output the conclusion without authorization page, if there is, then execute next step;Handwritten signature of author or tutor is detected in authorization page;Output authorization page automatic detection result.The application converts effective authorization page into image classification and target detection problem by automatic detection, and realizes the automatic detection of authorization page and the automatic discrimination of whether authorization page is effective by using image classification and target detection technology based on deep convolution network.

The present invention relates to the technical field of government information data security management, and specifically to a security management system and method for government information data based on big data. The present invention monitors the storage status parameters of various data storage terminals of government agencies, calculates a transfer pressure index, and compares it in combination with preset double thresholds. When the transfer pressure index exceeds the upper limit, a forced transfer instruction is automatically triggered, and semantic analysis is combined to identify transferable and non-transferable files. The use status evaluation value of the transferable files is further calculated, and the files are divided into positive demand files and non-demand files. The positive demand files are determined to be transmitted to the corresponding demand storage area according to the demand status evaluation value, and the non-demand files are determined to be transmitted to the corresponding important storage area according to the important status evaluation value. At the same time, the transmission of the files to be stored is encrypted and managed, thereby realizing dynamic adjustment and security management of government data and improving resource utilization and data security.

The teachings of the present disclosure include systems and / or methods for encoding digital data into a handwritten sample. An exemplary method includes accessing a predetermined vibration pattern stored in memory that corresponds to defined data, and vibrating a stylus based on the predetermined vibration pattern during creation of the handwritten sample to encode the defined data into the handwritten sample.

The invention provides a non-contact double-recording signature method and device based on end intelligence, and the method comprises the steps: loading a face recognition model and a gesture recognition model preset in end side equipment, and initializing a real-time video stream of a camera; face detection, living body judgment and facial feature comparison are carried out based on the real-time video stream, and an identity verification result is generated; synchronously recognizing a gesture of a user or an object moving track, matching a standard signature template through dynamic time warping, and generating a track consistency verification result; the identity verification result, the track consistency verification result and the signature track data are locally encrypted and stored, and are asynchronously uploaded to a server when the network is recovered; and through a timestamp alignment mechanism, audio and video streams in a signature process, an identity verification result and track data are synchronously recorded, encrypted and stored. According to the invention, the authenticity and security of signature are effectively improved, the risk of data leakage is reduced, and the system stability is enhanced.

The invention provides a title review process management method and system, and relates to the technical field of review process management, and the method comprises the steps: constructing a database for storing historical review features, obtaining to-be-reviewed data, extracting the features of the to-be-reviewed data to obtain to-be-reviewed features, and carrying out the preliminary review through a preset review rule to obtain a preliminary review result; if the preliminary review result conforms to the expectation, calculating a first similarity between the to-be-reviewed feature and a working class sub-feature in the historical review feature, and if the preliminary review result does not conform to the expectation, sending out a material doubt signal; and judging whether the first similarity is greater than a first preset similarity threshold and less than 1, if so, sending a material doubt signal, and otherwise, sending a material legal signal. The review efficiency can be improved.

A mail signature verification system employs a software that is configured to remove pixels of a digital photograph taken of a signature through an opaque security window. The mail piece can be signed and then the signature is covered with the signature cover having a security window therein. The signature reading station of a mail processing machine compresses the security window against a press plate and illuminates the signature with a light element while taking the digital photograph. The software then analyzes the digital photograph to remove pixels having a grey scale value below an upper grey scale threshold and remove pixels above a lower grey scale threshold, thereby rendering a signature with these darkest of pixels over the lightest of pixels.

A method and system for writable document forgery detection that includes encrypting a reference writing sample, compressing the reference writing sample, embedding the reference writing sample onto the writable document, receiving the writable document, the writable document including a handwritten segment, encrypting the handwritten segment of the received writable document, comparing the encrypted reference writing sample to the encrypted handwritten segment, and assessing a likelihood of the encrypted reference writing sample and the encrypted handwritten segment being written by a same person based on the comparison.