Chinese character structure and geometric space interactive writing method

Abstract

The invention discloses a Chinese character structure and geometric space interactive writing method, which relates to the technical field of interactive writing methods, and comprises the following steps: carrying out geometric classification processing on Chinese characters by adopting a structural analysis method to obtain a Chinese character type set comprising a left-right structure, an up-down structure and a surrounding structure; carrying out geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule base containing component proportion, gravity center alignment and stroke topological relation; performing structured storage processing on the obtained geometric constraint rule base by adopting a database construction method to obtain a dynamically-called Chinese character writing template database; the writing process of the user is monitored in real time, and a writing feature data set containing stroke coordinates, writing speed and pressure data is obtained; performing matching processing on the obtained writing feature data by adopting a dynamic grid generation method to obtain a self-adaptive reference grid corresponding to the currently written Chinese character; and carrying out deviation detection processing on the obtained self-adaptive reference grid by adopting a computer vision method.

Description

Technical Field

[0001] The present invention relates to the technical field of interactive writing methods, in particular to a method for interactive writing of Chinese character structures and geometric spaces. Background Art

[0002] Interactive writing methods are a teaching method that uses modern information technology to monitor and analyze the user's writing process in real time and provide immediate feedback. Therefore, how to use advanced technologies to improve the intelligence level of interactive writing methods is one of the current issues that need to be addressed.

[0003] In the field of interactive writing methods, existing automatic writing assessment methods perform poorly when dealing with different font styles and writing habits. In particular, their recognition and analysis capabilities for handwritten Chinese characters are limited, making it difficult to provide accurate feedback information. In addition, the writing rules of complex Chinese characters such as enclosing structures are relatively abstract, making it difficult for learners to grasp their essence through simple graphic tutorials. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an interactive writing method for Chinese character structure and geometric space to solve the problem that existing automatic writing evaluation methods perform poorly when dealing with different font styles and writing habits, especially the limited recognition and analysis capabilities of handwritten Chinese characters, making it difficult to provide accurate feedback information, and the writing rules of complex Chinese characters such as enclosing structures are relatively abstract, making it difficult for learners to grasp their essence through simple graphic tutorials.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for interactively writing Chinese character structure and geometric space, comprising:

[0008] The structural analysis method is used to geometrically classify Chinese characters, and a set of Chinese character types including left-right structure, top-bottom structure and surrounding structure is obtained.

[0009] Performing geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule library including component proportions, center of gravity alignment and stroke topological relationships;

[0010] The obtained geometric constraint rule base is structured and stored using a database construction method to obtain a dynamically callable Chinese character writing template database.

[0011] Monitor the user's writing process in real time to obtain a writing feature dataset containing stroke coordinates, writing speed, and pressure data;

[0012] A dynamic grid generation method is used to match the obtained writing feature data to obtain an adaptive reference grid corresponding to the currently written Chinese character;

[0013] Computer vision methods are used to detect deviations of the obtained adaptive reference grid, and writing deviation analysis results and real-time feedback instructions are obtained;

[0014] The transfer learning method is used to generate training tasks for the writing deviation analysis results, and a composite structure training system including difficulty grading and evaluation reports is obtained.

[0015] As a preferred embodiment of the interactive writing method of Chinese character structure and geometric space of the present invention, wherein: the structural analysis method is used to perform geometric classification processing on Chinese characters to obtain a set of Chinese character types including left-right structure, top-bottom structure and surrounding structure, the specific steps are as follows:

[0016] The image preprocessing algorithm is used to perform skeleton extraction on the original input Chinese character image data, and then the Chinese character skeleton graph SCG is obtained;

[0017] The skeleton graph obtained from the Chinese character skeleton graph (SCG) is used to automatically segment the structural components and obtain the geometric information of the position coordinates, direction vector and area size of each component;

[0018] The structural type determination formula is used to determine the structural type of each component's position coordinates, direction vectors, and area size geometric information to obtain the structural category label to which the Chinese character belongs.

[0019] The obtained structure type information is classified and stored, and the component information of the corresponding structure type is grouped and saved in a Chinese character structure type collection library;

[0020] Perform frequency statistics on the number of various structures in the Chinese character structure type collection library to obtain the distribution ratio information of the structure types;

[0021] The structural type distribution ratio information is used to prioritize the structural type frequency information to generate a structural complexity evaluation factor sequence, and the structural complexity information is sorted to generate a structural priority ranking list.

[0022] As a preferred embodiment of the interactive writing method of Chinese character structure and geometric space of the present invention, the geometric rule definition processing is performed on the obtained Chinese character type set to obtain a geometric constraint rule library including component proportions, center of gravity alignment and stroke topological relationships, and the specific steps are as follows:

[0023] A geometric analysis algorithm is used to analyze the structural component information of each Chinese character in the Chinese character type set to obtain the relative position relationship, area ratio and direction vector between each component;

[0024] Based on the relative position relationship, area ratio and direction vector, the center of gravity of each Chinese character is calculated using the center of gravity calculation formula to obtain the center of gravity coordinates of the Chinese character;

[0025] Based on the structural type of Chinese characters, the barycentric coordinates are calibrated using geometric constraints to obtain the calibrated barycentric coordinates.

[0026] The structural types of the Chinese characters include left-right, top-bottom, or surrounding;

[0027] The stroke topology relationship recognition algorithm is used to analyze and process the stroke connection graph (SCG) of Chinese characters to obtain the connection sequence and intersection information between each stroke;

[0028] Combining the obtained center-of-gravity coordinates and stroke topological relationships, a geometric constraint rule library is constructed for each structure type.

[0029] As a preferred solution of the interactive writing method of Chinese character structure and geometric space of the present invention, wherein: the geometric constraint rule library obtained by the database construction method is subjected to structured storage processing to obtain a dynamically callable Chinese character writing template database, the specific steps are as follows:

[0030] Each rule information in the geometric constraint rule library is divided into fields to obtain a standardized data structure including structure type, component ratio, center of gravity offset threshold and stroke topology relationship;

[0031] Define the table structure of the standardized rule record set and establish a relational database model for storing geometric constraint rules;

[0032] The data insertion interface is used to import rules into the relational database model, and all geometric constraint rules are grouped by structure type and written into the corresponding database table. The index optimization algorithm is then used to optimize the query performance of the imported database, generating a joint index structure based on structure type and component ratio.

[0033] The template mapping mechanism is used to map the geometric constraint rules in the database to a writing template, generating a writing template record containing structural parameters, reference grid configuration and stroke order constraints.

[0034] Classify and organize the writing template record collection to build a Chinese character writing template database that supports dynamic calling;

[0035] API interface encapsulation technology is used to encapsulate the Chinese character writing template database and generate a database access interface that can be called in real time by the writing process monitoring module.

[0036] As a preferred embodiment of the interactive writing method of Chinese character structure and geometric space of the present invention, the real-time monitoring of the user's writing process to obtain a writing feature dataset including stroke coordinates, writing speed and pressure data is carried out in the following specific steps:

[0037] The sensor is used to capture and process the signal generated by the user's writing tool to obtain the original writing trajectory data;

[0038] De-noising and smoothing are performed on the original writing trajectory data to obtain a clean stroke coordinate sequence;

[0039] Based on the stroke coordinate sequence, the timestamp synchronization technology is used to synchronize the timestamp of each coordinate point and calculate the time interval between adjacent coordinate points. and displacement vector

[0040] Based on time interval and displacement vector Estimate the writing speed and get the instantaneous writing speed v corresponding to each coordinate point ij ;

[0041] Combining the stroke coordinate sequence, instantaneous writing speed, and pressure data in the original writing trajectory, a comprehensive feature vector F containing stroke coordinates, writing speed, and pressure information is constructed;

[0042] The comprehensive feature vectors F of multiple strokes are aggregated to form a writing feature dataset for a single Chinese character or the entire writing task.

[0043] As a preferred solution of the interactive writing method of Chinese character structure and geometric space of the present invention, wherein: the dynamic grid generation method is used to match the obtained writing feature data to obtain an adaptive reference grid corresponding to the currently written Chinese character, and the specific steps are as follows:

[0044] Use the template retrieval interface to query the Chinese character writing template database and extract the corresponding writing template parameter set according to the current writing character encoding or structure type;

[0045] A coordinate normalization algorithm is used to normalize the stroke coordinate sequence in the writing feature dataset D to obtain a normalized coordinate point set to fit the standardized writing space.

[0046] A shape matching algorithm is used to calculate the similarity between the normalized coordinate point set and the template parameter set to obtain the structural similarity between the current written Chinese character and the template;

[0047] A structural type correction algorithm is used to dynamically adjust the original reference grid configuration information to generate an adaptive reference grid based on the current writing style.

[0048] The adaptive reference grid G′ is graphically rendered using visual rendering technology and is superimposed on the real-time trajectory written by the user.

[0049] As a preferred solution of the interactive writing method of Chinese character structure and geometric space of the present invention, wherein: the deviation detection processing of the obtained adaptive reference grid using a computer vision method to obtain the writing deviation analysis result and real-time feedback instruction is carried out, the specific steps are:

[0050] Grayscale and edge extraction are performed on the adaptive reference grid G′ and the user's actual writing trajectory image to obtain a binary contour image for subsequent comparison;

[0051] Perform overlapping area recognition processing on the grid contour image and the writing contour image to obtain a local offset map;

[0052] The error integral model is used to perform global deviation quantification on the offset map M(x,y), and then the overall deviation index of writing is obtained.

[0053] The direction consistency analysis algorithm is used to statistically process the vector directions in the offset map M(x,y) to obtain the writing direction deviation factor;

[0054] The pressure-speed correlation model was used to jointly analyze and process the writing speed and pressure data to obtain the dynamic writing stability index.

[0055] The overall deviation degree and direction deviation factor D f and stability index S d Perform weighted fusion processing to generate a comprehensive writing deviation score S c ;

[0056] Score for comprehensive writing deviation c Perform threshold determination processing and then generate a real-time feedback instruction set F including graphic prompts, voice feedback or vibration reminders c ;

[0057] During the processing, if S c <T1 is rated as "excellent", no obvious deviation, no feedback is required, if T1≤S c <T2 is rated as "good", and slight auxiliary prompts are provided. c ≥T2 was rated as “poor” and detailed corrective feedback was generated.

[0058] As a preferred embodiment of the interactive writing method of Chinese character structure and geometric space of the present invention, the transfer learning method is used to generate training tasks for the obtained writing deviation analysis results to obtain a composite structure training system including difficulty grading and evaluation reports. The specific steps are as follows:

[0059] The data collection mechanism was used to analyze the results of writing deviations. c Summarize and process the user-written feature dataset D and construct a basic dataset for subsequent model training

[0060] Screen the key features in the basic data set to determine the core feature set that affects writing quality;

[0061] Use pre-trained model loading technology to initialize the existing high-quality handwriting assessment model to obtain a baseline model with preliminary handwriting quality judgment capabilities;

[0062] Fine-tune the baseline model based on the current user's writing dataset and optimize the model parameters to adapt to the individual user's writing habits to generate a personalized evaluation model M. p ;

[0063] Personalized evaluation model M p The output writing deviation scores are classified and divided into training tasks of different difficulty levels according to the score range

[0064] The training tasks of different difficulty levels are combined with the user's historical writing records for comprehensive analysis and processing to generate a personal evaluation report containing progress trends, common error types and improvement suggestions.

[0065] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the interactive writing method of Chinese character structure and geometric space as described in the first aspect of the present invention is implemented.

[0066] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for interactive writing of Chinese character structure and geometric space as described in the first aspect of the present invention is implemented.

[0067] The beneficial effects of the present invention are as follows: by modeling the geometric features such as the position, area, and direction of structural components, combined with center of gravity calculation and stroke connection analysis, the geometric expression of Chinese character writing standards is realized, making writing teaching more logical and scientific, and improving user learning efficiency; through field division, table structure design, index optimization and API encapsulation, efficient organization and rapid calling of geometric rules are achieved, which ensures that the method can still maintain efficient operation under large-scale data, and provides a stable data foundation for subsequent adaptive reference grid generation and deviation detection; through sensor acquisition, trajectory smoothing processing and multi-parameter fusion modeling, digital capture of user writing behavior is achieved, which improves the interactivity and instant feedback capability of the method, and enhances user experience and teaching effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0069] Figure 1 This is a flow chart of the interactive writing method of Chinese character structure and geometric space in Example 1. DETAILED DESCRIPTION

[0070] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0072] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0073] Example 1, with reference to Figure 1 , which is the first embodiment of the present invention, provides a method for interactively writing Chinese character structure and geometric space, comprising the following steps:

[0074] S1. Use structural analysis to geometrically classify Chinese characters and obtain a set of Chinese character types including left-right structure, top-bottom structure, and surrounding structure;

[0075] Furthermore, an image preprocessing algorithm is used to perform skeleton extraction on the original input Chinese character image data, and then a Chinese character skeleton graph SCG is obtained;

[0076] The skeleton graph obtained from the Chinese character skeleton graph (SCG) is used to automatically segment the structural components and obtain the geometric information of the position coordinates, direction vector and area size of each component;

[0077] The structural type determination formula is used to process the geometric information of the position coordinates, direction vectors and area size of each component to determine the structural type, and the structural category label to which the Chinese character belongs is obtained. The expression is:

[0078]

[0079] Among them, T is the final judgment structure type, L represents the left-right structure, U represents the up-down structure, E represents the surrounding structure, D(p i ,p j ) represents the Euclidean distance between the i-th component and the j-th component, A(p i ,p j ) represents the relative area matching weight function between two components, and n is the total number of structural components contained in the current Chinese character;

[0080] The obtained structure type information is classified and stored, and the component information of the corresponding structure type is grouped and saved in a Chinese character structure type collection library;

[0081] Perform frequency statistics on the number of various structures in the Chinese character structure type collection library to obtain the distribution ratio information of the structure types;

[0082] The structural type distribution ratio information is used to prioritize the structural type frequency information to generate a structural complexity evaluation factor sequence, and the structural complexity information is sorted to generate a structural priority ranking list;

[0083] It should be noted that the above-mentioned structural classification method is not only applicable to the recognition and processing of printed Chinese characters, but also to handwritten Chinese characters. It has good robustness and adaptability. By introducing the structural type determination formula, it realizes the transformation from visual features to mathematical expressions, so that the distinction between different structural types has a clear quantitative basis, providing a standardized input basis for subsequent writing standard modeling.

[0084] S2. performing geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule library including component proportions, center of gravity alignment, and stroke topological relationships;

[0085] Furthermore, a geometric analysis algorithm is used to parse the structural component information of each Chinese character in the Chinese character type set to obtain the relative position relationship, area ratio and direction vector between each component. The expression is:

[0086]

[0087]

[0088] Where i and j represent the i-th and j-th components respectively;

[0089] Based on the relative position relationship, area ratio and direction vector, the center of gravity of each Chinese character is calculated using the center of gravity calculation formula to obtain the center of gravity coordinates of the Chinese character;

[0090] Based on the structural type of Chinese characters, the barycentric coordinates are calibrated using geometric constraints to obtain the calibrated barycentric coordinates.

[0091] The structural types of Chinese characters include left-right, top-bottom, or surrounding;

[0092] The stroke topology relationship recognition algorithm is used to analyze and process the stroke connection graph (SCG) of Chinese characters to obtain the connection sequence and intersection information between each stroke;

[0093] Combining the obtained center of gravity coordinates and stroke topological relationships, a geometric constraint rule library for each structure type is constructed;

[0094] It should be noted that the geometric constraint rule library constructed in this step integrates the spatial distribution characteristics of structural components and the stroke connection relationship, and can effectively support writing teaching and evaluation in various font styles. Through the center of gravity offset calibration mechanism and topological relationship analysis, it enhances the control ability of complex structure writing stability, thereby improving the overall intelligent guidance level of the method.

[0095] S3. Using a database construction method to perform structured storage processing on the obtained geometric constraint rule library to obtain a dynamically callable Chinese character writing template database;

[0096] Furthermore, each rule information in the geometric constraint rule library is divided into fields to obtain a standardized data structure including structure type, component ratio, center of gravity offset threshold and stroke topology relationship;

[0097] Define the table structure of the standardized rule record set and establish a relational database model for storing geometric constraint rules;

[0098] The data insertion interface is used to import rules into the relational database model, and all geometric constraint rules are grouped by structure type and written into the corresponding database table. The index optimization algorithm is then used to optimize the query performance of the imported database, generating a joint index structure based on structure type and component ratio.

[0099] The template mapping mechanism is used to map the geometric constraint rules in the database to a writing template, generating a writing template record containing structural parameters, reference grid configuration and stroke order constraints.

[0100] Classify and organize the writing template record collection to build a Chinese character writing template database that supports dynamic calling;

[0101] The API interface encapsulation technology is used to encapsulate the Chinese character writing template database and generate a database access interface that can be called in real time by the writing process monitoring module;

[0102] It should be noted that the use of a structured database management method enables geometric rules to be efficiently stored and called according to multi-dimensional conditions such as structural type and component proportion, which greatly improves the scalability and response efficiency of the method. Through API interface encapsulation, it ensures that the template database can seamlessly connect with other functional modules to form a complete teaching feedback closed loop, which is suitable for large-scale personalized teaching scenario deployment.

[0103] S4, monitoring the user's writing process in real time to obtain a writing feature dataset including stroke coordinates, writing speed, and pressure data;

[0104] Furthermore, a sensor is used to capture and process the signal generated by the user's writing tool to obtain the original writing trajectory data;

[0105] De-noising and smoothing are performed on the original writing trajectory data to obtain a clean stroke coordinate sequence;

[0106] Based on the stroke coordinate sequence, the timestamp synchronization technology is used to synchronize the timestamp of each coordinate point and calculate the time interval between adjacent coordinate points. and displacement vector The expression is:

[0107]

[0108] Among them, i and j represent the i-th and j-th coordinate points respectively;

[0109] Based on time interval and displacement vector Estimate the writing speed and get the instantaneous writing speed v corresponding to each coordinate point ij , the expression is:

[0110]

[0111] Among them, v ij is the instantaneous writing speed;

[0112] Combining the stroke coordinate sequence, instantaneous writing speed, and pressure data in the original writing trajectory, a comprehensive feature vector F containing stroke coordinates, writing speed, and pressure information is constructed;

[0113] Aggregate the comprehensive feature vectors F of multiple strokes to form a writing feature dataset for a single Chinese character or the entire writing task;

[0114] It should be noted that this step constructs a multi-dimensional writing feature system by real-time collection and comprehensive modeling of the trajectory, speed and pressure data during the user's writing process, providing high-quality data support for subsequent deviation detection and personalized training. The timestamp synchronization mechanism and feature vector construction method adopted ensure the temporal consistency and structural integrity of the data, and enhance the accuracy of model analysis.

[0115] S5, using a dynamic grid generation method to perform matching processing on the obtained writing feature data to obtain an adaptive reference grid corresponding to the currently written Chinese character;

[0116] Furthermore, a template retrieval interface is used to query the Chinese character writing template database and extract the corresponding writing template parameter set according to the current writing character encoding or structure type;

[0117] A coordinate normalization algorithm is used to normalize the stroke coordinate sequence in the writing feature dataset D to obtain a normalized coordinate point set to fit the standardized writing space.

[0118] The shape matching algorithm is used to calculate the similarity between the normalized coordinate point set and the template parameter set to obtain the structural similarity between the current written Chinese character and the template. The expression is:

[0119]

[0120] Among them, S is the structural similarity, p∈N is the coordinate point written by the current user, q∈M is the corresponding standard grid point in the template, and d avg is the average distance between all point pairs, |N| is the number of currently written points;

[0121] The structural type correction algorithm is used to dynamically adjust the original reference grid configuration information to generate an adaptive reference grid based on the current writing style. The expression is:

[0122]

[0123] Among them, G is the original template grid, G is the center distribution of the grid actually written by the user, S is the structural similarity,

[0124] userα adjustment factor, G′ is the final adaptive reference grid;

[0125] The adaptive reference grid G′ is rendered using visual rendering technology and displayed overlaid on the real-time trajectory of the user's writing.

[0126] It should be noted that the generation mechanism of the adaptive reference grid fully considers the differences between the user's individual writing habits and the standard template. Through normalized coordinate mapping and shape matching algorithm, it realizes the transition from static template to dynamic guidance. Combined with visual rendering technology, the reference grid is superimposed on the actual writing trajectory, which helps to improve the user's visual perception ability and writing correction efficiency.

[0127] S6. Using computer vision methods to perform deviation detection processing on the obtained adaptive reference grid, obtaining writing deviation analysis results and real-time feedback instructions;

[0128] Furthermore, the adaptive reference grid G′ and the user’s actual writing trajectory image are grayscaled and edge extracted to obtain a binary contour image for subsequent comparison;

[0129] Perform overlapping area recognition processing on the grid contour image and the writing contour image to obtain a local offset map;

[0130] The error integral model is used to perform global deviation quantification on the offset map M(x,y), and then the overall deviation index of writing is obtained.

[0131] The direction consistency analysis algorithm is used to perform statistical processing on the vector direction in the offset map M(x,y) to obtain the writing direction deviation factor, which is expressed as:

[0132]

[0133] in, is the direction vector of the reference grid at that point, The direction vector of the user-written path at that point, D f is the direction deviation factor;

[0134] The pressure-speed correlation model is used to jointly analyze and process the writing speed and pressure data to obtain the dynamic writing stability index, which is expressed as:

[0135]

[0136] Among them, m is the total number of coordinate points, σ v is the speed mean and standard deviation, σ p is the mean and standard deviation of pressure, S d is the dynamic stability index;

[0137] The overall deviation degree and direction deviation factor D f and stability index S d Perform weighted fusion processing to generate a comprehensive writing deviation score S c ;

[0138] Score for comprehensive writing deviation c Perform threshold determination processing and then generate a real-time feedback instruction set F including graphic prompts, voice feedback or vibration reminders c ;

[0139] During the processing, if S c <T1 is rated as "excellent", no obvious deviation, no feedback is required, if T1≤S c <T2 is rated as "good", and slight auxiliary prompts are provided. c ≥T2 was rated as “poor” and detailed corrective feedback was generated;

[0140] It should be noted that this step uses computer vision technology to perform high-precision comparison of the writing trajectory with the reference grid, and combines multiple indicators such as error integral, direction deviation factor and dynamic stability index to achieve a comprehensive evaluation of the writing quality. The comprehensive score generated by the weighted fusion mechanism can accurately reflect the user's writing performance and drive a variety of real-time feedback command outputs, significantly improving the interactivity and practicality of teaching.

[0141] S7. Using a transfer learning method to generate training tasks for the obtained handwriting deviation analysis results, thereby obtaining a composite structure training system including difficulty grading and evaluation reports;

[0142] Furthermore, a data collection mechanism is used to analyze the writing deviation results S c Summarize and process the user-written feature dataset D and construct a basic dataset for subsequent model training

[0143] Screen the key features in the basic data set to determine the core feature set that affects writing quality;

[0144] Use pre-trained model loading technology to initialize the existing high-quality handwriting assessment model to obtain a baseline model with preliminary handwriting quality judgment capabilities;

[0145] Fine-tune the baseline model based on the current user's writing dataset and optimize the model parameters to adapt to the individual user's writing habits to generate a personalized evaluation model M. p ;

[0146] Personalized evaluation model M p The output writing deviation scores are classified and divided into training tasks of different difficulty levels according to the score range

[0147] Comprehensively analyze and process the training tasks of different difficulty levels combined with the user's historical writing records to generate a personal evaluation report that includes progress trends, common error types, and improvement suggestions;

[0148] It should be noted that the application of transfer learning strategy enables the method to quickly adapt to new users under limited sample conditions, avoiding the waste of resources and response delays caused by training from scratch. Through the difficulty grading and personal assessment report generation mechanism, a personalized training path recommendation system is formed to help users accurately identify their own problems and continuously optimize their writing skills, truly realizing the teaching concept of teaching students in accordance with their aptitude.

[0149] This embodiment also provides a computer device, which is suitable for the interactive writing method of Chinese character structure and geometric space, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the interactive writing method of Chinese character structure and geometric space proposed in the above embodiment.

[0150] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0151] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for interactively writing Chinese character structure and geometric space proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0152] In summary, the present invention realizes the geometric expression of Chinese character writing standards by modeling the geometric features such as the position, area, and direction of structural components, combining center of gravity calculation and stroke connection analysis, making writing teaching more logical and scientific, and improving user learning efficiency. Through field division, table structure design, index optimization and API encapsulation, efficient organization and rapid calling of geometric rules are achieved, which ensures that the method can still maintain efficient operation under large-scale data, and provides a stable data foundation for subsequent adaptive reference grid generation and deviation detection. Through sensor acquisition, trajectory smoothing processing and multi-parameter fusion modeling, digital capture of user writing behavior is achieved, which improves the interactivity and instant feedback capability of the method, and enhances user experience and teaching effect.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for interactively writing Chinese character structure and geometric space, characterized by: include: The structural analysis method is used to geometrically classify Chinese characters, and a set of Chinese character types including left-right structure, top-bottom structure and surrounding structure is obtained. Performing geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule library including component proportions, center of gravity alignment and stroke topological relationships; The obtained geometric constraint rule base is structured and stored using a database construction method to obtain a dynamically callable Chinese character writing template database. Monitor the user's writing process in real time to obtain a writing feature dataset containing stroke coordinates, writing speed, and pressure data; A dynamic grid generation method is used to match the obtained writing feature data to obtain an adaptive reference grid corresponding to the currently written Chinese character; Computer vision methods are used to detect deviations of the obtained adaptive reference grid, and writing deviation analysis results and real-time feedback instructions are obtained; The transfer learning method is used to generate training tasks for the writing deviation analysis results, and a composite structure training system including difficulty grading and evaluation reports is obtained.

2. The interactive writing method of Chinese character structure and geometric space according to claim 1, characterized in that: The structural analysis method is used to perform geometric classification processing on Chinese characters to obtain a set of Chinese character types including left-right structure, top-bottom structure and surrounding structure. The specific steps are as follows: The image preprocessing algorithm is used to perform skeleton extraction on the original input Chinese character image data, and then the Chinese character skeleton graph SCG is obtained; The skeleton graph obtained from the Chinese character skeleton graph (SCG) is used to automatically segment the structural components and obtain the geometric information of the position coordinates, direction vector and area size of each component; The structural type determination formula is used to determine the structural type of each component's position coordinates, direction vectors, and area size geometric information to obtain the structural category label to which the Chinese character belongs. The obtained structure type information is classified and stored, and the component information of the corresponding structure type is grouped and saved in a Chinese character structure type collection library; Perform frequency statistics on the number of various structures in the Chinese character structure type collection library to obtain the distribution ratio information of the structure types; The structural type distribution ratio information is used to prioritize the structural type frequency information to generate a structural complexity evaluation factor sequence, and the structural complexity information is sorted to generate a structural priority ranking list.

3. The interactive writing method of Chinese character structure and geometric space according to claim 2, characterized in that: The geometric rule definition processing is performed on the obtained Chinese character type set to obtain a geometric constraint rule library including component proportions, center of gravity alignment and stroke topological relationships. The specific steps are as follows: A geometric analysis algorithm is used to analyze the structural component information of each Chinese character in the Chinese character type set to obtain the relative position relationship, area ratio and direction vector between each component; Based on the relative position relationship, area ratio and direction vector, the center of gravity of each Chinese character is calculated using the center of gravity calculation formula to obtain the center of gravity coordinates of the Chinese character; Based on the structural type of Chinese characters, the barycentric coordinates are calibrated using geometric constraints to obtain the calibrated barycentric coordinates. The structural types of the Chinese characters include left-right, top-bottom, or surrounding; The stroke topology relationship recognition algorithm is used to analyze and process the stroke connection graph (SCG) of Chinese characters to obtain the connection sequence and intersection information between each stroke; Combining the obtained center-of-gravity coordinates and stroke topological relationships, a geometric constraint rule library is constructed for each structure type.

4. The interactive writing method of Chinese character structure and geometric space according to claim 3, characterized in that: The database construction method is used to perform structured storage processing on the obtained geometric constraint rule library to obtain a dynamically callable Chinese character writing template database. The specific steps are: Each rule information in the geometric constraint rule library is divided into fields to obtain a standardized data structure including structure type, component ratio, center of gravity offset threshold and stroke topology relationship; Define the table structure of the standardized rule record set and establish a relational database model for storing geometric constraint rules; The data insertion interface is used to import rules into the relational database model, and all geometric constraint rules are grouped by structure type and written into the corresponding database table. The index optimization algorithm is then used to optimize the query performance of the imported database, generating a joint index structure based on structure type and component ratio. The template mapping mechanism is used to map the geometric constraint rules in the database to a writing template, generating a writing template record containing structural parameters, reference grid configuration and stroke order constraints. Classify and organize the writing template record collection to build a Chinese character writing template database that supports dynamic calling; API interface encapsulation technology is used to encapsulate the Chinese character writing template database and generate a database access interface that can be called in real time by the writing process monitoring module.

5. The interactive writing method of Chinese character structure and geometric space according to claim 4, characterized in that: The real-time monitoring of the user's writing process to obtain a writing feature dataset including stroke coordinates, writing speed and pressure data is carried out in the following specific steps: The sensor is used to capture and process the signal generated by the user's writing tool to obtain the original writing trajectory data; De-noising and smoothing are performed on the original writing trajectory data to obtain a clean stroke coordinate sequence; Based on the stroke coordinate sequence, the timestamp synchronization technology is used to synchronize the timestamp of each coordinate point and calculate the time interval between adjacent coordinate points. and displacement vector Based on time interval and displacement vector Estimate the writing speed and get the instantaneous writing speed corresponding to each coordinate point Combining the stroke coordinate sequence, instantaneous writing speed, and pressure data in the original writing trajectory, a comprehensive feature vector F containing stroke coordinates, writing speed, and pressure information is constructed; The comprehensive feature vectors F of multiple strokes are aggregated to form a writing feature dataset for a single Chinese character or the entire writing task.

6. The interactive writing method of Chinese character structure and geometric space according to claim 5, characterized in that: The dynamic grid generation method is used to match the obtained writing feature data to obtain an adaptive reference grid corresponding to the currently written Chinese character. The specific steps are: Use the template retrieval interface to query the Chinese character writing template database and extract the corresponding writing template parameter set according to the current writing character encoding or structure type; A coordinate normalization algorithm is used to normalize the stroke coordinate sequence in the writing feature dataset D to obtain a normalized coordinate point set to fit the standardized writing space. A shape matching algorithm is used to calculate the similarity between the normalized coordinate point set and the template parameter set to obtain the structural similarity between the current written Chinese character and the template; A structural type correction algorithm is used to dynamically adjust the original reference grid configuration information to generate an adaptive reference grid based on the current writing style. Among them, G is the original template grid, The grid center distribution of the user's actual writing, S structure similarity, α adjustment factor, G ′ The final adaptive reference mesh; Adaptive reference grid G ​​is visualized using rendering technology. ′ Perform graphic drawing processing and overlay it on the real-time trajectory written by the user.

7. The interactive writing method of Chinese character structure and geometric space according to claim 6, characterized in that: The computer vision method is used to perform deviation detection processing on the obtained adaptive reference grid to obtain writing deviation analysis results and real-time feedback instructions. The specific steps are: For the adaptive reference grid G ′ Perform grayscale conversion and edge extraction on the user's actual writing trajectory image to obtain a binary contour image for subsequent comparison; Perform overlapping area recognition processing on the grid contour image and the writing contour image to obtain a local offset map; The error integral model is used to perform global deviation quantification on the offset map M(x,y), and then the overall deviation index of writing is obtained. The direction consistency analysis algorithm is used to statistically process the vector directions in the offset map M(x,y) to obtain the writing direction deviation factor; The pressure-speed correlation model was used to jointly analyze and process the writing speed and pressure data to obtain the dynamic writing stability index. The overall deviation degree and direction deviation factor D f and stability index S d Perform weighted fusion processing to generate a comprehensive writing deviation score S c ; Score for comprehensive writing deviation c Perform threshold determination processing and then generate a real-time feedback instruction set F including graphic prompts, voice feedback or vibration reminders c ; During the processing, if S c <T1 is rated as "excellent", no obvious deviation, no feedback is required, if T1≤S c <T2 is rated as "good", providing slight auxiliary prompts. If S c ≥T2 is rated as "poor" and detailed corrective feedback is generated.

8. The interactive writing method of Chinese character structure and geometric space according to claim 7, characterized in that: The transfer learning method is used to generate training tasks for the obtained writing deviation analysis results to obtain a composite structure training system including difficulty grading and evaluation reports. The specific steps are as follows: The data collection mechanism was used to analyze the results of writing deviations. c Summarize and process the user-written feature dataset D and construct a basic dataset for subsequent model training Screen the key features in the basic data set to determine the core feature set that affects writing quality; Use pre-trained model loading technology to initialize the existing high-quality handwriting assessment model to obtain a baseline model with preliminary handwriting quality judgment capabilities; Fine-tune the baseline model based on the current user's writing dataset and optimize the model parameters to adapt to the individual user's writing habits to generate a personalized evaluation model M. p ; Personalized evaluation model M p The output writing deviation scores are classified and divided into training tasks of different difficulty levels according to the score range The training tasks of different difficulty levels are combined with the user's historical writing records for comprehensive analysis and processing to generate a personal evaluation report containing progress trends, common error types and improvement suggestions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the Chinese character structure and geometric space interactive writing method described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the Chinese character structure and geometric space interactive writing method described in any one of claims 1 to 8 are implemented.

Images