67 results about "Handwriting recognition" patented technology

The invention discloses a subjective question scoring and evaluation system based on SOLO classification, and belongs to the field of machine paper marking. In order to solve the defect that the thinking structure level of students cannot be reflected due to the fact that only attention scores are concerned in traditional machine paper marking of subjective questions, an SOLO classification theory is introduced, keyword-logic-thinking hierarchy mapping is constructed through multiple rounds of cue word engineering and directional fine adjustment in combination with a large language model, whole-section semantic analysis and hierarchical score assignment are achieved, and the thinking structure level of the students cannot be reflected. Generating a traceable report; oCR handwriting recognition is fused to form a closed-loop automatic system to adapt to an offline scene; custom configuration of teachers is supported, multi-dimensional learning condition visualization data is generated, and teaching is assisted. The method is applied to the field of intelligent education and has the advantages of automation, flexibility and the like.

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 invention relates to the technical field of teaching, and discloses a handwriting recognition method, which comprises the following steps that S1, Chinese character writing specifications are dynamically displayed through AR, a user participates in practice in real time, and the practice is analyzed in real time; s2, a multi-sensory collaborative teaching practice module which synchronously integrates visual, auditory and tactile feedbacks; s3, forming a review strategy based on an intelligent memory enhancement mechanism of the user behavior data; s4, gamifying an interactive interface, and supporting real-time stroke detection and error correction; and S5, a personalized learning path generation module adapted to the learning progress and error mode of the user. According to the method, the writing track of the user is analyzed in real time through the double-branch convolutional neural network, three-level error classification is output, and AR dynamic feedback is combined, so that the stroke sequence detection rate is 92%, the font analysis delay is less than 20 ms, and the error correction efficiency is improved; the multi-sensory collaboration module dynamically distributes visual, auditory and tactile weights by using an attention mechanism to adapt to different learning preferences, so that the memory retention rate is improved, and the environment adaptive capacity is enhanced.

The invention relates to a handwritten character recognition method and device, equipment, a storage medium and a chip, and the method comprises the steps: obtaining handwritten character information which comprises the OCR information of a target subject in an education scene; calling a handwriting recognition model to carry out subject character recognition on the handwritten character information to obtain a corresponding recognition result; the handwriting recognition model comprises m path architectures, m is a positive integer, and the target path architecture is used for performing subject character recognition on the handwritten character information; the target path architecture is a path architecture selected from m path architectures at least based on one of an input mode, a task type, input complexity and input characteristics of the handwritten character information. In this way, the technical problems that in the prior art, a universal model is redundant, the response time is long, and the accuracy of character recognition and the robustness of a model algorithm cannot be guaranteed can be solved.

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 invention discloses an end-to-end volley handwritten letter recognition method based on a millimeter wave radar, and belongs to the field of radar data processing. According to the invention, complete radar original data is used to identify volley handwritten letters, and a multi-modal optimization target of a millimeter wave radar end-to-end handwriting identification method based on writing mode guidance is provided. A multi-modal handwriting recognition model for millimeter wave radar original signals and two-dimensional handwritten letter stroke trajectory data is provided, and a targeted neural network training strategy is provided for the optimization target and the model architecture. According to the method, priori knowledge is fused in a training process, so that the extraction complexity of abstract features by a network is reduced, the training convergence of a radar original data feature extractor is accelerated, and the recognition accuracy of the radar original data feature extractor in a complex handwritten letter recognition task is improved.

The invention discloses an optical character recognition method and system for mixing a handwritten form and a printed form, relates to the technical field of computer vision and document analysis, and adopts a handwritten form OCR (Optical Character Recognition) model to carry out handwritten form area detection on an input document image, and the input document image is divided into a printed body internal handwritten form, an independent handwritten paragraph and a table handwritten form. Through a two-channel model architecture and a fine-grained classification strategy, the problems of low detection precision and poor model generalization are solved, and efficient and lightweight mixed text analysis is realized. The improved YOLOv12-L model is based on spatial layout and writing feature classification, can cope with diversified document scenes, and reduces recognition failures caused by scene differences. Through classification optimization, the core problems of false detection, semantic fracture of complex typesetting, table interference and the like in a mixed scene can be solved in a targeted manner, and scene self-adaption of end-to-end handwriting recognition is achieved.

The embodiment of the invention provides a handwriting system and a handwriting recognition method. The handwriting system comprises a writing surface, a handwriting pen and an electronic device, the writing surface comprises at least one optical identification code, and each optical identification code corresponds to one coordinate; the handwriting pen is used for determining a written target content according to a plurality of target optical identification codes passed by a user when the user uses the handwriting pen to write on the writing surface; the electronic device comprises a display screen, and the display screen is used for displaying the target content. According to the scheme, the writing track when the user writes the characters can be identified through the optical identification code on the writing surface, and the writing track can assist in identifying the written characters, so that the accuracy of handwriting identification is improved.

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.

To suppress inconvenience caused by forced conversion into a predetermined character type.SOLUTION: A display device according to the present invention which performs character recognition on a handwritten input using a dictionary and converts the handwritten input into a predetermined character type, comprises: a handwriting recognition control unit that calculates accuracy information relating to a recognition result for each handwritten character; a handwriting recognition dictionary control unit that uses a first dictionary to recognize the handwritten character and convert it into a first character string of a first character type when the accuracy information exceeds a predetermined threshold, and uses a second dictionary including a second character type different from the first character type predetermined in the first dictionary to recognize the handwritten character and convert it into a second character string of the second character type when the accuracy information is less than or equal to the predetermined threshold; and a display control unit that displays, in a predetermined input field in which forced conversion is performed, the first character string of the first character type recognized using the first dictionary and the second character string of the second character type recognized using the second dictionary.SELECTED DRAWING: Figure 9

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 invention discloses a handwritten character recognition method, a related device, equipment and a storage medium, and the method comprises the steps: carrying out the character recognition based on a handwriting track of a target object, and obtaining a first text; wherein the first text comprises first characters respectively represented by each handwritten character in the handwritten track; performing similarity comparison in a character library of the target object based on the handwritten characters to obtain second characters represented by the handwritten characters; wherein the character library comprises real characters respectively represented by a plurality of reference characters, and the reference characters are handwritten characters historically written by the target object; and replacing the first character represented by the corresponding handwritten character in the first text based on the second character represented by the handwritten character to obtain a second text. According to the scheme, the accuracy of handwriting recognition can be improved.

The invention provides a blackboard-writing display method, electronic equipment and a storage medium, the method is applied to the technical field of interactive tablets, and the method comprises the following steps: displaying a whiteboard application interface; in response to the identification instruction for the whiteboard application interface, identifying the original handwriting on the whiteboard application interface to obtain a handwriting identification result; acquiring voice information recorded in the writing process, and recognizing the voice information to obtain a voice recognition result; generating structured blackboard writing content according to the handwriting recognition result and the voice recognition result; wherein the writing style of the structured blackboard writing content is the same as that of the original handwriting; and displaying the structured blackboard writing content on the whiteboard application interface. According to the method, structuring of the blackboard-writing content is achieved, the readability, the constitutive property and the semantic integrity of the blackboard-writing content are improved, and a user can understand and reuse the blackboard-writing content conveniently.

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 provides a handwriting display method and device, electronic equipment and a storage medium, and the method comprises the steps: displaying a handwriting track corresponding to a handwriting operation in a handwriting region of a handwriting screen under the condition that the handwriting operation is detected; and performing handwriting recognition on the handwriting track to obtain a handwriting recognition result, and displaying the handwriting recognition result in a display area of the handwriting screen. By displaying the different display areas and the handwriting areas on the same screen, a user can synchronously view the handwriting content and the recognition result thereof in the same visual range without frequent switching among the different areas, so that the operation process is simplified, and the conciseness of an interface and the user interaction efficiency are improved; the size of the display area is increased along with the increase of the handwriting recognition result, and the size of the handwriting area is reduced along with the increase of the handwriting recognition result, so that the space of the handwriting area and the display area is reasonably distributed, the natural writing habit of the user is met, and efficient handwriting interaction experience is provided for the user.

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.

The present application provides a method, device and system for early warning of Alzheimer's disease risk based on handwriting recognition, which relates to the field of health information processing technology. The method includes: receiving handwriting recognition data from the Bluetooth pen in real time while the assessor is filling out a paper diagnostic scale with a handwriting pen; calculating the test subject's diagnostic scale score based on the handwriting recognition data and determining the test subject's first Alzheimer's disease risk level; using an Alzheimer's disease risk prediction model to determine the test subject's second Alzheimer's disease risk level, and obtaining the test subject's Alzheimer's disease risk warning result data. The present application can be applied to test subjects such as the elderly who are inconvenient to use electronic scales, and can ensure the efficiency and automation of paper scale data processing, can effectively improve the accuracy and reliability of Alzheimer's disease risk warning, and thus can better achieve auxiliary diagnosis of Alzheimer's disease for the test subject.

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.

The application relates to a design method of an artificial intelligence-based official document approval intelligent transcription system, and belongs to the field of computer software.The application establishes a handwriting recognition algorithm model aiming at handwriting signature content recognition, first acquires a handwriting annotation area image, provides input for handwriting content recognition, extracts handwriting approval content, avoids manual transcription and input work, and reduces the difficulty and quantity of manual recognition work.In order to improve the correctness of handwriting approval final acquisition content, a content error correction link is added, a content error correction algorithm model is established, error detection is carried out from two aspects of word granularity and character granularity through word segmentation, an optimal correction word is obtained through a language model, error correction is completed, and the final accuracy of handwriting content recognition is greatly improved;through template engine technology, a transcription template is constructed, and a document is automatically generated, manual arrangement and document input work is saved, and the file circulation speed is greatly 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 invention discloses a Transform-based offline handwriting data generation method, which comprises the following steps of: respectively capturing long-range context information of a text content image and a handwriting style image based on a Transform encoder to obtain a style feature sequence and a content feature sequence; the style feature sequence and the content feature sequence are continuously coded based on a Transform decoder, and depth feature coupling is carried out; inputting the coupling features into a CNN decoder to directly obtain a handwriting generation image with a specific handwriting style; and inputting the generated handwriting image into a character-level style discriminator and a stroke-level style discriminator for fine-grained style and content consistency discrimination. According to the method, the global writing style and the coding local style can be captured, the content and the style are decoupled in a fine-grained manner, the high-quality handwriting image is generated, and the accuracy of handwriting recognition is effectively improved.