41 results about "Handwriting recognition" patented technology

This invention relates to systems and methods for real-time dynamic classification and orchestration of test automation components in distributed DevOps environments. The system features an Auto Identify Automation (AIA) engine that leverages supervised learning, Multi-Modal Artificial Intelligence (AI), and Generative AI technologies. It includes a Smart Scenario Designer interface that allows users to author test scenarios using handwriting and voice inputs, which are processed in real-time by AI-driven handwriting recognition, voice recognition, and Natural Language Processing (NLP). The system dynamically suggests relevant automated components via a smart bubble pane, facilitating rapid scenario creation. The architecture is tool-agnostic and scalable, with a Shared Workbench Engine that supports real-time collaboration and conflict resolution. The system continuously adapts and improves, ensuring that the automation suite remains consistent, up-to-date, and aligned with evolving software requirements, enabling efficient and user-friendly management of complex test automation processes.

The present disclosure relates to systems, software, and computer-implemented methods for automatically identifying handwritten tips. An example method includes obtaining an image of a receipt, where the image includes a reference region in a pre-set format, a plurality of category identifiers, and at least one set of handwritten characters. The reference region and the plurality of category identifiers can be printed. The plurality of category identifiers can be located at pre-set positions relative to the reference region. The method further includes obtaining optical character recognition (OCR) information of the image and identifying the reference region from the OCR information based on the pre-set format of the reference region. The method further includes determining a plurality of amounts based on the OCR information and the reference region, where at least one of the plurality of amounts is associated with a tip category.

The invention discloses a dictation content generation interaction method based on a large model, and relates to the technical field of educational information processing, and the method comprises the steps: 1, unifying characters, pictures or audios into a reference text, reserving pose anchor points, segmenting the reference text into target units according to semantics, and numbering the target units; 2, combining with the writing duration of a learner portrait prediction unit, and generating a time axis script including a broadcast text, a starting moment, reading duration, pause and a prompt position; step 3, broadcasting and collecting pen leaving, coverage and a password according to a script, determining an actual completion moment, adjusting a next unit to stop according to an error, and recording an execution log; step 4, obtaining an answer image, generating an error cause label through pose matrix correction, handwriting recognition and alignment, and superposing an annotation on a paper surface or a screen; and step 5, writing back the learner portrait and pause according to the execution log and the error cause, generating a next round of time axis script, and finally forming a rhythm closed-loop continuous updating mechanism.

The application relates to the technical field of pattern recognition, and provides a handwriting recognition method and device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining handwriting data to be recognized, the handwriting data comprising matched stroke trajectories and handwriting images; based on a trajectory encoder, trajectory features of the stroke trajectories are extracted, and based on an image encoder, image features of the handwriting images are extracted, the trajectory encoder and the image encoder being obtained through differentiated training strategies; based on a dual-modal decoder, differentiated information between the image features and the trajectory features is applied to fuse the trajectory features and the image features, and handwriting recognition is performed based on the fused features. The handwriting recognition method and device, the electronic equipment and the storage medium provided by the application can significantly improve the recognition capability in a dual-modal data mismatching scene.

We provide the system. [Solution] A means of automatically generating treatment policies and care plans by analyzing patient information using generative artificial intelligence, A method for converting visit records into digital data using handwriting recognition technology and centrally managing the information, A means of automatically generating answers to questions outside of one's area of ​​expertise based on analyzed information, We protect information in secure data centers and implement measures to ensure the strict management of highly confidential information. A system that includes this.

The invention discloses a whole-process data perception and comprehensive evaluation method and system for intelligent homework accompanying, and belongs to the technical field of intelligent education and human-computer interaction. The method comprises the following steps: before homework, generating a dynamic homework plan based on a homework list, student historical data and a real-time state; during operation, academic, behavior and emotion data are collected through a multi-modal sensor and are processed under an end-cloud architecture; in the writing process, handwriting recognition is combined with a knowledge graph to realize real-time discovery and attribution of wrong questions; after operation, a comprehensive evaluation report is generated based on the multi-dimensional evaluation model, feedback is output through a natural language generation module, and the report is stored after being subjected to credible evidence storage processing. The system realizes objective monitoring and personalized evaluation of the whole process of student homework through cooperation of dynamic adaptive planning, real-time error attribution, multi-modal perception fusion and block chain evidence storage, and has the advantages of improving accompanying quality, enhancing report credibility and facilitating long-term tracking.

The application is a real-time handwriting recognition method based on deep neural network, aiming to solve the problems of insufficient accuracy, poor real-time performance and poor adaptability to complex writing style in electronic whiteboard handwriting character recognition. The method first collects handwriting trajectory sample data through electronic whiteboard sensors and infrared cameras, and after missing value processing, outlier processing and format conversion, the trajectory coordinates are denoised and normalized. Secondly, enhanced samples are generated by pixel shift, 10-dimensional features such as direction angle, speed and normalized pen tip pressure are extracted, and particle swarm optimization algorithm is used to select the optimal feature subset and splice it into time sequence feature vector. Finally, a 3-layer LSTM network is used to learn the time sequence dependence of the feature vector, and a full connection layer and a softmax function are used to output the probability of 62 classes of characters, and the classification recognition is completed. The method realizes efficient and accurate recognition of electronic whiteboard handwriting characters, improves the real-time performance and reliability of handwriting interaction, and has strong engineering application value.

This invention discloses an electronic pen handwriting recognition method and system based on ultrasonic transmission trajectory. The method includes: receiving handwriting trajectory data transmitted by an electronic pen via ultrasound and received by a sensor array, the trajectory data including spatial coordinate information and corresponding timestamps; performing time alignment and resampling processing on the trajectory data, and obtaining a continuous trajectory sequence by interpolation to compensate for missing trajectory points; constructing a multidimensional dynamic feature sequence containing spatial location features, kinematic features, and trajectory geometric features based on the continuous trajectory sequence; inputting the multidimensional dynamic feature sequence into a two-branch Transformer feature encoding network to extract trajectory spatial structure features and motion change features respectively, and performing feature fusion through a cross-branch attention mechanism; inputting the fused features into a sequence decoder to generate a corresponding character sequence, thereby achieving text recognition of the continuous handwriting trajectory. This invention is applicable to online handwriting recognition scenarios based on ultrasonically positioned electronic pens.

The application relates to the technical field of character recognition, and provides a general-scene character recognition data automatic generation method and system, which comprises the following steps: collecting and establishing a part of character recognition data set A; using a deep learning-based image enhancement method to generate printed data set B; using an adversarial generative network (GAN) to migrate the style of the data set A to the data set B, and expanding to obtain data set C; using a CTC-based character accurate positioning image enhancement method to obtain D and E; and proportionally mixing the data sets A, B, C, D and E for training, wherein the mixed training adopts a feature-based data enhancement method to enrich sample diversity. The application can solve the problem that, in handwriting character recognition, due to the continuous writing of handwriting characters, the handwriting characters are more diverse, meanwhile, diverse fonts and open data sets are lacking, and handwriting recognition is difficult to achieve the same amazing effect as that of typewritten characters, and high-quality and evenly-distributed data are generated through image enhancement, semantic understanding, GAN and other technologies.

The embodiment of the invention discloses a conference information display method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring current display information in a multi-person conference scene; obtaining first handwriting information corresponding to the current wearing user, second handwriting information corresponding to the current display information, display video information corresponding to the current display information and conference audio information associated with the current display information; identifying the first handwriting information, and determining a display information processing instruction corresponding to the first handwriting information; based on the display information processing instruction, controlling a device worn by the current wearing user to perform marking processing on the current display information; inputting the second handwriting information, the display video information and the conference audio information into a writing prediction model to obtain a writing prediction result; and displaying the writing prediction result in the current display information. According to the technical scheme, the handwriting recognition and prediction efficiency and accuracy can be improved in a complex multi-person conference scene.

The invention belongs to the technical field of image processing, and relates to a handwriting recognition method and device, a medium and electronic equipment. The method comprises the following steps: acquiring a target character image, and performing feature extraction processing on the target character image by using a capsule network to obtain an initial feature map; performing feature weighting processing on the initial feature map to obtain a weighted feature map, and inputting the weighted feature map into a pre-trained first recurrent neural network to obtain a first feature; and inputting the initial feature map into a pre-trained second recurrent neural network to obtain a second feature, and determining an identification result of the target character image according to the first feature and the second feature. The method gives play to the advantage that the capsule network is good at processing the position and attitude change of an object in the image, can understand the hierarchical structure of the image, can remember the information in a long sequence by using the recurrent neural network, is suitable for understanding the time sequence characteristics of characters, has certain advantages in predicting scribble fonts, and has a good application prospect. Therefore, the accuracy and robustness of character recognition are improved.

The application provides a handwriting recognition method and system based on ultrasonic waves, a mobile terminal and a server, the application combines ultrasonic waves with handwriting recognition, proposes a new input method, and through machine learning technology combining CTC loss and LSTM, realizes real-time recognition of handwriting input; the ultrasonic waves after being carried by the autocorrelation sequence improve the robustness to environmental changes, and in a cross-environment experiment, a high letter recognition accuracy is achieved; the machine learning model has good segmentation ability for the interval between words, can support continuous handwriting input recognition, solves the problem that long sequence input is difficult to segment in actual application, and partially solves the cross-environment problem faced by passive sensing, and shows great potential of ultrasonic wave passive recognition in commercial fields.

A method for providing recommendations based on handwritten data in a handwritten document is disclosed. The method includes fetching the handwritten data corresponding to strokes made by a user in the handwritten document on a screen of an electronic device. The method also includes recognizing text associated with the fetched handwritten data by employing a handwriting recognition technique. The method further includes identifying keywords in the recognized text and determining recommendations based on the identified keywords using a named entity linking and / or a knowledge graph. The recommendations include handwritten documents stored in a database and / or weblinks. The method also includes ranking the determined recommendations based on a similarity score of each of the determined one or more recommendations with respect to the identified one or more keywords and rendering the ranked recommendations to the user.

The invention discloses a signature identification method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: acquiring a to-be-detected image; for each signature in the to-be-detected image, the signature is split into a plurality of stroke units, and the stroke units are minimum units forming the signature; extracting handwriting features of each stroke unit, the handwriting features including geometric features and contour features; a handwriting recognition model is adopted to process and analyze the handwriting features, a plurality of matching probabilities are output, each matching probability corresponds to one target object, and the matching probabilities are used for indicating the probability that the signatures belong to the target objects; and determining the target object corresponding to the maximum matching probability in the plurality of matching probabilities as the affiliation object of the signature in the to-be-detected image. The technical problem of low accuracy of an identification result caused by neglecting detail information of a signature in a handwriting identification method in related technologies is solved.

The present invention relates to an electronic whiteboard with improved handwriting recognition for an optical touch panel, and an operation method thereof. An objective of the present invention is to provide an optical electronic whiteboard and an operation method thereof, in which an object and a touch position of the object can be precisely identified by overcoming physical limitations, such as a size of a light-emitting element in a touch screen using an optical ray touch panel. To this end, according to the present invention, in the electronic whiteboard which recognizes coordinates of an object from sensing data sensed through an optical processing unit, and comprises a screen for visually displaying content generated on the basis of the recognized coordinates of the object: the optical processing unit comprises an optical transmission unit including a plurality of light-emitting elements installed on the screen, and an optical reception unit including a plurality of light detection sensors installed at positions corresponding to the optical transmission unit so as to receive optical rays emitted from the light-emitting elements; the optical transmission unit includes the light-emitting elements for emitting optical rays, and an optical path converter installed on the front portion of the light-emitting elements so as to convert the direction of the optical rays emitted from the light-emitting elements; and the optical path converter is a prism sheet having a trapezoidal prism pattern formed thereon, and converts the direction of the optical rays emitted from the light-emitting elements into a plurality of directions.

The invention provides a handwriting acquisition method, system and terminal, and the method comprises the steps: obtaining the data information of a pen body in the writing process, the data information comprises the acceleration, angular velocity and magnetic field intensity of the pen body in the writing process; performing compensation processing on the angular velocity according to the acceleration and the magnetic field intensity; constructing an earth fixed connection coordinate system, a pen body coordinate system and a writing coordinate system; according to the compensated angular velocity, a conversion quaternion for conversion among the pen body coordinate system, the earth fixed connection coordinate system and the writing coordinate system is updated to obtain an updated conversion quaternion; according to a preset decomposition algorithm and the updated quaternion, carrying out decomposition processing on handwriting information when the pen body writes on the writing coordinate system so as to obtain a pen point displacement track of the pen body; and mapping the pen point displacement track to a preset display device. According to the scheme, the accuracy of handwriting recognition and the degree of freedom of writing based on the intelligent pen can be improved.

An electronic device for processing handwritten input is disclosed, which includes a touch screen, a processor operatively connected with the touch screen, and a memory operatively connected with the processor, wherein the memory stores instructions which, when executed, cause the processor to control the electronic device to: perform handwriting recognition on a first handwritten input of a user displayed on the touch screen to convert the first handwritten input into text; identify at least one of attributes or features of the first handwritten input; apply at least one of the identified attributes or features to the converted text; and replace the first handwritten input with the text to which at least one of the identified attributes or features has been applied (hereinafter referred to as first rich text) in response to a request for conversion of the first handwritten input.

The embodiment of the invention provides a handwritten character recognition method and system based on handwriting manifold field modeling, and the method comprises the steps: carrying out the handwriting form enhancement of an input image, generating a handwriting skeleton diagram representing a stroke center track through a deformable convolutional network, and constructing a handwriting manifold field containing the direction, width, pressure and speed based on the handwriting skeleton diagram. Secondly, setting a dual-channel recognition structure, performing vision and handwriting structure recognition through different models based on the input image and handwriting related features to obtain two character probability distributions, and generating fusion weight weighted fusion according to local quality information and feature difference; and inputting the final recognition result into the lightweight text error correction model, and correcting the low-confidence character. The handwriting structure and the writing dynamics characteristics are accurately captured through the handwriting manifold field, and the two-channel dynamic fusion mechanism is combined to adapt to different scenes, so that the difficulty of scribble character and continuous handwriting recognition is effectively broken through, the complex background interference is resisted, and the accuracy of handwriting character recognition and the system robustness are improved.

This invention concerns a system for recognizing online handwriting comprising: a handwriting instrument including a body extending longitudinally between a first end and a second end, the first end having a writing tip which is able to write on a support, the handwriting instrument further including a module comprising at least one motion sensor configured to acquire motion data on the handwriting of a user when the user is writing a characters sequence with the handwriting instrument, a calculating unit communicating with the at least one motion sensor and configured to analyze the motion data by a machine learning model trained in a multitask way such that it is capable of performing at least two tasks at the same time, the machine learning model being configured to deliver as an output the characters sequence which was written by the user with the handwriting instrument.

The invention relates to the technical field of image recognition, in particular to an OCR (Optical Character Recognition) method, device and equipment based on a large model and a storage medium, which can directly parse structured semantic information with clear logic from an original image through the powerful image-text joint modeling capability of a visual language large model in combination with visual and text reasoning, and improve the recognition efficiency. And manual secondary processing on the intermediate result is not needed, so that the automation degree and efficiency of processing are greatly improved. Besides, the handwritten character recognition sub-model is introduced to enhance handwritten recognition and face various signature styles, the sub-model can combine visual features and text semantics to carry out comprehensive analysis, recognition accuracy is remarkably improved, the situation of recognition failure caused by the problems of irregular strokes, continuous strokes and the like is avoided, effective extraction of key handwritten information is guaranteed, and user experience is improved. The problem that end-to-end understanding from an original image to structured semantic information is difficult to realize in a traditional OCR scheme is solved, and the technical problem that the handwriting recognition effect is poor is solved.

The application provides a handwriting recording and playing method, device, equipment and storage medium, the handwriting recording method converts the collected track data into handwriting instructions containing instruction type, coordinate information and handwriting timestamp by initializing global time scale, and writes the handwriting instructions into instruction file. The audio data is recorded synchronously, and the audio timestamp is marked and written into the corresponding audio file. The handwriting playing method reads the instruction file and audio file, filters the handwriting instructions based on the global time scale to redraw the handwriting picture, and synchronously plays the audio data. The recording and playing based on the handwriting instruction stream has the advantages of low space occupation and no loss of definition compared with the video-based scheme, is convenient for fast distribution under mobile network, and can avoid handwriting edge blur caused by video compression. Moreover, since the handwriting coordinate information is stored, additional operations such as handwriting recognition or adaptive rendering under different resolutions based on handwriting information are supported.

Disclosed are an electronic device for supplementing note content on the basis of text identified from a handwriting input, and an operation method therefor. The electronic device may comprise a memory for storing instructions, and at least one processor for executing the instructions, wherein when the at least one processor individually and / or collectively executes the instructions, the instructions cause the electronic device to: acquire, through a note application, a handwriting image based on a handwriting input; input the handwriting image to a handwriting recognition model to generate first text data corresponding to the handwriting input; generate first metadata on the basis of the first text data and first supplementary data input in relation to the handwriting image; generate second metadata through a search within the electronic device on the basis of the first metadata; and generate second text data in which the first text data has been supplemented, on the basis of the first text data and the second metadata.

Examples described herein generally relate to systems and methods for handwriting recognition. In an example, a computing device may receive input corresponding to a handwritten word and apply first recognition model to the input. The first recognition model may be configured to determine a first confidence level of a first portion of the input is greater than a second confidence level of a second portion of the input. The computing device may also apply a second recognition model to the input, wherein the second recognition model is different from the first recognition model and combine results of the first recognition model and the second recognition model to determine a list of candidate words. The computing device may also output one or more candidate words from the list of candidate words.

Methods, systems, and computer-readable media related to a technique for providing handwriting input functionality on a user device. A handwriting recognition module is trained to have a repertoire comprising multiple non-overlapping scripts and capable of recognizing tens of thousands of characters using a single handwriting recognition model. The handwriting input module provides real-time, stroke-order and stroke-direction independent handwriting recognition for multi-character handwriting input. In particular, real-time, stroke-order and stroke-direction independent handwriting recognition is provided for multi-character, or sentence level Chinese handwriting recognition. User interfaces for providing the handwriting input functionality are also disclosed.

The invention relates to the technical field of image recognition, and discloses a multi-route OCR recognition system based on multi-recognition-target classification, and the system comprises an image feature extraction module which is used for obtaining a shallow feature tensor; the route distribution module is used for generating a route control mask according to the local gradient distribution; the parallel recognition module comprises a printing recognition unit and a handwriting recognition unit, the routing distribution module extracts gradient tensors by utilizing a directional derivative operator, determines anisotropic response characteristic values according to gradient variance ratios, constructs routing control masks accordingly, distributes the characteristic tensors through element-by-element multiplication operation, and outputs the characteristic tensors to the printing recognition unit. According to the method, the stroke topological distribution features are established by using the anisotropic response feature values, so that semantic decoupling of heterogeneous features is completed before deep recognition, feature space aliasing is eliminated, the composite layout recognition precision is improved, and computing power resource allocation is coordinated.

The application relates to the computer field and discloses a method and device for updating a document based on annotation content and a display device. The method comprises the following steps: in response to an annotation operation, displaying an annotation window above a display window of a target document; the target document can be displayed through the annotation window; in response to a touch operation in the annotation window, displaying a touch track of the touch operation in the annotation window and determining annotation content according to the touch track; determining an update strategy in response to a selection operation in the annotation window; or determining the update strategy according to the touch track or the annotation content; and updating the target document according to the update strategy and the annotation content. A transparent window is created above the document, handwriting input is performed on the transparent window, handwriting input content is automatically recognized by using a handwriting recognition algorithm, the original document is updated according to the recognized content at a determined update position, and the conference record is conveniently arranged and displayed, and the conference efficiency is improved.

The invention relates to the technical field of education and artificial intelligence, in particular to a Chinese character visual recognition training system based on deep learning. Comprising a handwriting recognition module used for receiving handwriting sequence data input by a user and performing recognition processing on the handwriting sequence data to obtain handwriting recognition confidence; the user state monitoring module is used for obtaining a cognitive load index for quantifying the cognitive load state of the user; the decision arbitration module is used for obtaining the final intervention intensity, and the final intervention intensity decouples the direct binding relation between the handwriting recognition confidence and the teaching intervention intensity; and the dynamic intervention module is used for executing teaching feedback of intensity, granularity and frequency adaptation to the user according to the final intervention intensity. According to the method, a vicious negative feedback cycle that strong intervention is caused by a poor user state and the user state is poorer due to the strong intervention is broken through, and collaborative improvement of teaching effectiveness and system robustness is realized.

A method and an electronic device for processing data using generative artificial intelligence (AI), and a non-transitory computer-readable storage medium are provided. The method is executed by the electronic device and includes: obtaining and displaying corresponding word information based on a startup input operation performed by a user; obtaining and displaying corresponding handwriting recognition information based on a handwriting input operation performed by the user; generating weight adjustment ratio information based on a generated content weight adjustment operation performed by the user; generating an AI data processing mode based on a generated content selection input operation performed by the user; generating an AI-generated content by using generative AI based on the word information, the handwriting recognition information, the weight adjustment ratio information, and the AI data processing mode; and displaying the AI-generated content.

The application claims a faiss-based online handwritten signature handwriting retrieval method and system, and relates to the field of electronic signature handwriting recognition. By using big data and deep learning technology to train and extract handwriting features with identity recognition, the electronic signature data features are vectorized, and then a large-scale handwriting feature library is established. For the handwriting data to be identified, the handwriting feature vector is extracted, and then compared with the feature vector in the feature library. The vector similarity is calculated, and the final topN handwriting retrieval result is obtained according to the similarity ranking.

The application relates to an online signature authentication method integrating point features, local features and global features, and relates to the technical field of electronic signature handwriting authentication. The point features of a sample signature and a to-be-tested signature are subjected to time dynamic programming to obtain the point feature similarity of the sample signature and the to-be-tested signature, the local feature similarity of the sample signature and the to-be-tested signature is calculated based on LNPS features, the global feature similarity of the sample signature and the to-be-tested signature is calculated according to the Euclidean distance, the comprehensive similarity of the sample signature and the to-be-tested signature is calculated according to the point feature similarity, the local feature similarity and the global feature similarity, and the signature is authenticated according to the comprehensive similarity. The authentication result is more real and reliable, and the accuracy and precision of judgment in the handwriting recognition and authentication process are improved.