Calligraphy practice auxiliary system based on image analysis

Abstract

The invention discloses a calligraphy practice auxiliary system based on image analysis, relates to the technical field of image analysis, and solves various practical problems encountered by calligraphy beginners in regular script calligraphy practice in the prior art. In order to solve the problems that in the prior art, a calligraphy practice evaluation model is constructed through three core indexes including the error character rate, the standard writing deviation coefficient and the character pattern deviation value, the position deviation and the character pattern structure in a writing boundary grid are accurately mastered, and evaluation and adjustment cannot be made for each practice, an all-around evaluation result can be provided for a user, and the accuracy of the calligraphy practice is improved. In addition, the evaluation model can quantify the calligraphy practice effect of the user and compare the practice result of this time with a historical result, the user is helped to intuitively understand progress or backstepping of himself / herself, and the mode of feeding back the practice effect through the quantified data is beneficial to improving the calligraphy practice effect of the user. And the user can be helped to see own progress in a long-term exercise process.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a calligraphy practice auxiliary system based on image analysis. Background Art

[0002] In traditional calligraphy learning, beginners mainly rely on the demonstration and step-by-step guidance of calligraphy teachers, or learn by copying calligraphy copybooks. With the development of science and technology, some modern technologies, such as intelligent calligraphy teaching software and digital calligraphy pens, are gradually being applied to calligraphy teaching. These technologies can record the writing process, provide basic feedback information, and help calligraphy learners to train themselves.

[0003] Most of the existing calligraphy auxiliary tools are unable to comprehensively evaluate the position of calligraphy glyphs in the grid, stroke proportions, structural layout, etc. Especially in the study of regular script calligraphy, the standardization requirements of the font are high, and the use of the grid is very important, which provides beginners with a reference for standardized writing areas and proportions. However, current calligraphy auxiliary tools can usually only provide an overall deviation judgment of the glyphs, but cannot be detailed to the accuracy of each character in the grid. This technical limitation makes it difficult for calligraphy learners to accurately grasp the basic structure and stroke proportions of Chinese characters, and cannot help beginners effectively improve their calligraphy level. Most of the existing technologies can only give a simple "good" or "bad" evaluation, lacking detailed analysis results. This rough feedback method cannot meet the needs of beginners for fine practice.

[0004] Existing technologies cannot solve the many practical problems that beginners encounter in regular script calligraphy practice, including positional deviations in the writing grid, accurate grasp of the glyph structure, and inability to evaluate and adjust each practice. These problems limit beginners' progress and make the calligraphy learning process lack systematic and data-based guidance. Therefore, how to design an intelligent system based on image analysis that can evaluate the Chinese character writing of calligraphy practitioners in real time, give accurate feedback, and provide targeted improvement suggestions based on the evaluation results has become an important direction for the development of current calligraphy teaching technology. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a calligraphy practice auxiliary system based on image analysis, which solves the problems in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a calligraphy practice auxiliary system based on image analysis, comprising: An image acquisition module, used for acquiring images of the writer practicing on the handwriting paper through a camera; An image preprocessing module is used to preprocess the image in each grid of the acquired screen image to obtain a processed target image, and send it to a subsequent image deviation analysis module; The image deviation analysis module is used to analyze the target image, extract each grid of the target image, and compare it with the corresponding standard grid to determine the writing deviation coefficient of each analyzed grid, and finally obtain the standard writing deviation coefficient of this writing paper, and identify the wrong word grids at the same time, and calculate the wrong word rate of the writing paper according to the total number of grids of the writing paper; The image glyph analysis module uses an edge detection algorithm to extract the glyph outline of the Chinese character in the analyzed grid, determines the starting point of the glyph outline, overlaps it with the starting point of the standard outline by establishing a rectangular coordinate system, and identifies key feature points, and determines the standard writing glyph deviation value of the writing paper by calculation; The calligraphy practice evaluation module is used to establish a calligraphy practice evaluation model based on three core indicators: the error rate, the standard writing deviation coefficient of this writing paper, and the standard writing font deviation value of this writing paper. By comparing the calligraphy practice evaluation results of this writing paper with the historical evaluation results, it is determined whether there is progress and areas that need correction.

[0007] Preferably, in the image acquisition module, the camera shoots in the following manner: by placing a handwriting paper of a fixed specification at a reference position, the camera shoots the paper and captures the image of the practiced calligraphy.

[0008] Preferably, the reference position is determined as follows: A two-dimensional rectangular coordinate system is established in the shooting area, and the center of the camera's field of view is used as the origin O of the coordinate system. The horizontal direction of the field of view is the x-axis, and the vertical direction is the y-axis. The field of view of the camera is a rectangular area with a width of W and a length of H. Set the size of the fixed-size handwriting paper to A×B, and there is an m×n boundary grid on the handwriting paper; Align the center of the handwriting paper with the center of the camera's field of view to get the lower left corner P of the handwriting paper. left Coordinates: ; Then in P left Mark the coordinate point as the reference position; By setting the camera's field of view larger than the size of the handwriting paper: W ≥ A, H ≥ B; If the camera's field of view cannot completely cover the handwriting paper, you need to adjust the distance D between the camera and the handwriting paper to increase the field of view. The camera's field of view is proportional to the distance D: in, and are the horizontal and vertical viewing angles of the camera, respectively; Adjust D to ensure that the field of view covers the writing paper.

[0009] Preferably, in the image preprocessing module, the preprocessing methods include: Grayscale processing is performed on the image in each grid of the acquired screen image to convert the color image into a grayscale image; Then, the grayscale image in each grid is binarized by setting the threshold using Otsu's method.

[0010] Preferably, in the image deviation analysis module, the specific analysis method is: S1: Determine the total number of pixels z in the grid; S2: Select m×n grids on the target image from left to right and from top to bottom as the grids to be analyzed; S3: Identify the Chinese character in the analyzed grid, and find the standard grid corresponding to the Chinese character from the database, wherein the standard grid contains the standard writing of the Chinese character; The specific method of recognizing the Chinese characters in the analyzed grid is as follows: recognizing the Chinese characters in the analyzed grid by optical character recognition technology, marking the Chinese characters as wrong characters if they cannot be recognized; then recognizing the Chinese characters in the next analyzed grid; S4: Compare and analyze the Chinese characters in the analyzed grid with the standard grid to determine the writing deviation coefficient Ci of the analyzed grid; S5: repeating steps S3-S4, determining the number Q1 of specific wrong character grids in the m×n grids, and the writing deviation coefficient Ci of each analyzed grid in the non-wrong character grids; Determine the error rate F by the formula F=Q1 / (m×n); At the same time, the number of writing deviation coefficients Ci of the analyzed grids in the non-mistaken character grids is determined to be Q2, and Q2=(m×n)-Q1; the average value of the writing deviation coefficients Ci of Q2 analyzed grids in the non-mistaken character grids is taken as the standard writing deviation coefficient Cp of this writing paper.

[0011] Preferably, in step S4, the specific method of determining the writing deviation coefficient Ci of the analyzed grid is: S41: First determine the number of pixels H1 occupied by Chinese characters in the standard grid, and mark the pixels in the grid as Mb i , mark the pixels in the analyzed grid as Mc i ; where 1≤i≤z; S42: Then determine the number k of pixels with the value of 0 that are repeated in the Chinese characters in the standard grid and the analyzed grid, and determine the number j of pixels that do not match, wherein the pixels with the value of 0 in the analyzed grid corresponding to the pixels with the value of 255 in the standard grid are the pixels that do not match; S43: Determine the number of pixels k whose repeated values ​​are 0, and calculate r1=k / H1, where r1 is the boundary grid writing matching rate; Determine the number of mismatched pixels j, and calculate r2=j / H1, where r2 is the boundary grid writing misalignment rate; S44: The writing deviation coefficient Ci of the analyzed grid is calculated according to r1 and r2 by the following formula: .

[0012] Preferably, in the image glyph analysis module, the standard writing glyph deviation value of the writing paper is determined in the following manner: AS1: extract the target glyph outline of the Chinese character in the analyzed boundary by using an edge detection algorithm, and determine the starting point of the target glyph outline; The starting point is the position where the first stroke of the glyph outline begins; The starting point is overlapped with the starting point of the standard outline of the Chinese character, and a rectangular coordinate system is established with the point as the origin Oe, wherein the standard outline is the outline of the standard writing method of the Chinese character; AS2: Then obtain the coordinates of the turning point, intersection point and end point of the strokes of the standard outline of the Chinese character and the coordinates of the turning point, intersection point and end point of the strokes of the target glyph outline; The turning points and intersection points in the glyph are detected by using the Harris corner point detection method, and the end point is the end position of the last stroke of the glyph outline; AS3: Calculate the deviation value Li between the target glyph outline and the standard glyph outline through the inflection point, intersection point and end point of the matched strokes: Where N is the number of feature points, which is the total number of turning points and intersection points in the glyph detected by Harris corner detection plus the end points; ( )and( ) are the feature point coordinates of the target glyph outline and the standard glyph outline respectively; AS4: Then take the average value of the deviation values ​​Li of Q2 analyzed grids as the standard writing font deviation value Lp of this writing paper.

[0013] Preferably, in the calligraphy practice evaluation module, the calligraphy practice evaluation model is embodied by the following model formula: Among them, Stotal is the model result score, w1, w2 and w3 are the weight coefficients of the error rate, standard writing deviation coefficient and character shape deviation value, and w1+w2+w3=1 is satisfied.

[0014] The present invention provides a calligraphy practice auxiliary system based on image analysis. Compared with the prior art, it has the following beneficial effects: (1) The present invention can reduce noise and unnecessary information interference through standardized pre-processing, graying and binarization, thereby ensuring that the acquired image data is clear and visible. Combined with OCR technology to recognize characters, the present invention can effectively identify wrong characters. This precise analysis allows the practitioner to obtain objective feedback, which helps the practitioner to make targeted improvements. (2) The present invention helps users understand their mistakes and deviations after practice by analyzing the deviation of written images and character shapes. The image deviation analysis module analyzes the deviation of each character from the standard character shape and gives a deviation coefficient. By calculating the error rate and deviation value, users can clearly know which character is written incorrectly or has a large deviation. (3) The present invention constructs an evaluation model for calligraphy practice through three core indicators: character error rate, standard writing deviation coefficient and character shape deviation value. This systematic evaluation method can provide users with a comprehensive evaluation result, so that they can not only know where they are wrong in the practice process, but also understand the overall quality of their writing. In addition, the evaluation model can quantify the user's calligraphy practice effect and compare the current practice results with historical results to help users intuitively understand their progress or regression. This method of feedback on practice effects through quantitative data can help users see their own progress in the long-term practice process and form positive incentives, so that they can persist in practice and improve their overall calligraphy level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the accompanying drawings.

[0016] Figure 1 This is a structural framework diagram of a calligraphy practice auxiliary system based on image analysis according to the present invention; Figure 2 This is a processing flow chart of an image deviation analysis module of a calligraphy practice auxiliary system based on image analysis according to the present invention; Figure 3 This is a processing flow chart of an image character shape analysis module of a calligraphy practice auxiliary system based on image analysis according to the present invention; DETAILED DESCRIPTION The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1 See also Figure 1-Figure 3 The present invention provides a calligraphy practice auxiliary system based on image analysis, including an image acquisition module, an image preprocessing module, an image deviation analysis module, an image glyph analysis module and a calligraphy practice evaluation module: An image acquisition module, used for acquiring images of the writer practicing on the handwriting paper through a camera; Specifically, the handwriting paper is distributed through grids, and each character corresponds to a grid; The camera takes pictures in the following way: by placing a handwriting paper of a fixed size at a reference position, the camera takes pictures of the paper and collects the image of the calligraphy that has been practiced; The reference position is determined by establishing a two-dimensional rectangular coordinate system in the shooting area, with the center of the camera's field of view as the origin of the coordinate system O (0, 0), the horizontal direction of the field of view as the x-axis, the vertical direction as the y-axis, and the camera's field of view as a rectangular area with a width of W and a length of H. Set the size of the fixed-size handwriting paper to A×B (width A, length B), and there is an m×n boundary grid on the handwriting paper; Align the center of the handwriting paper with the center of the camera's field of view to get the lower left corner P of the handwriting paper. left Coordinates: Therefore, the handwriting paper is placed in the same position for each shot to capture the screen image; Then in P left Mark the coordinate point as the reference position so that the position will not shift each time the handwriting paper is placed; In order to ensure that the camera can fully cover the entire handwriting paper, the camera's field of view must be larger than the size of the handwriting paper, that is: W ≥ A, H ≥ B; If the camera's field of view cannot completely cover the handwriting paper, you need to adjust the distance D between the camera and the handwriting paper to increase the field of view. The camera's field of view is proportional to the distance D: in, and are the horizontal and vertical viewing angles of the camera, respectively; Adjust D to ensure that the field of view covers the handwriting paper; An image preprocessing module is used to preprocess the image in each grid of the acquired screen image to obtain a processed target image, and send it to a subsequent image deviation analysis module; Preprocessing methods include: Grayscale the image in each grid of the collected screen image, convert the color image into a grayscale image, and reduce unnecessary information interference; Then, the grayscale image in each grid is binarized by setting the threshold using Otsu's method. The steps of Otsu's method are as follows: Calculate the grayscale histogram: count the number of pixels at each grayscale level in the image to get the histogram of the image; Calculate the inter-class variance: By traversing all possible thresholds, calculate the variance of the foreground (stroke) and background under each threshold; the larger the inter-class variance, the better the separation effect of the foreground and background.

[0018] Determine the optimal threshold: select the threshold that maximizes the inter-class variance as the binarization threshold; Apply a threshold: Set pixels in the image that are less than or equal to the threshold to black (0), and pixels that are greater than the threshold to white (255); The binarization is performed using the automatically calculated optimal threshold value, which can provide clear and accurate stroke images for calligraphy practice; The image deviation analysis module is used to analyze the target image, extract each grid of the target image, and compare it with the corresponding standard grid to determine the writing deviation coefficient of each analyzed grid, and finally obtain the standard writing deviation coefficient of this writing paper, and identify the wrong word grids at the same time, and calculate the wrong word rate of the writing paper according to the total number of grids of the writing paper; See also Figure 2 , the specific analysis method is: S1: Determine the total number of pixels z in the grid; S2: Select m×n grids on the target image from left to right and from top to bottom as the grids to be analyzed; S3: Identify the Chinese character in the analyzed grid, and find the standard grid corresponding to the Chinese character from the database, wherein the standard grid contains the standard writing of the Chinese character; The specific method of identifying the Chinese characters in the analyzed grid is as follows: the Chinese characters in the analyzed grid are identified by the optical character recognition (OCR) technology in the prior art, and when the characters cannot be identified, they are marked as wrong characters in the grid; then the Chinese characters in the next analyzed grid are identified; S4: Compare and analyze the Chinese characters in the analyzed grid with the standard grid to determine the writing deviation coefficient Ci of the analyzed grid; The specific methods of comparative analysis are as follows: S41: First determine the number of pixels H1 occupied by Chinese characters in the standard grid, and mark the pixels in the grid as Mb i , mark the pixels in the analyzed grid as Mc i ; where 1≤i≤z; Specifically, Mb i With Mc i One to one correspondence; S42: Then determine the number k of pixels with the value of 0 that are repeated in the standard grid and the Chinese characters in the analyzed grid, and determine the number j of pixels that do not match, wherein the pixels with the value of 0 in the analyzed grid corresponding to the pixels with the value of 255 in the standard grid are the pixels that do not match; S43: Determine the number of pixels k whose repeated values ​​are 0, and calculate r1=k / H1, where r1 is the boundary grid writing matching rate; Determine the number of mismatched pixels j, and calculate r2=j / H1, where r2 is the boundary grid writing misalignment rate; It should be noted that after step S3, there are two situations for the analysis and judgment of S4. The first situation is that the boundary grid writing matching rate is high and the boundary grid writing misalignment rate is low. The second situation is that the boundary grid writing matching rate is low and the boundary grid writing misalignment rate is high. S44: The writing deviation coefficient Ci of the analyzed grid is calculated according to r1 and r2 by the following formula: S5: Repeat steps S3-S4 to determine the number Q1 of specific wrong character grids in the m×n grids, and the writing deviation coefficient Ci of each analyzed grid in the non-wrong character grids Determine the error rate F by the formula F=Q1 / (m×n); At the same time, the number of writing deviation coefficients Ci of the analyzed grids in the non-wrong character grids is determined to be Q2, and Q2=(m×n)-Q1; the average value of the writing deviation coefficients Ci of Q2 analyzed grids in the non-wrong character grids is taken as the standard writing deviation coefficient Cp of the writing paper this time; The image glyph analysis module uses an edge detection algorithm to extract the glyph outline of the Chinese character in the analyzed grid, determines the starting point of the glyph outline, and coincides it with the starting point of the standard outline by establishing a rectangular coordinate system, and uses Harris corner point detection to identify key feature points, and determines the standard writing glyph deviation value of the writing paper by calculation; The method for determining the standard writing font deviation value of the writing paper is as follows: AS1: extract the target glyph outline of the Chinese character in the analyzed boundary by using an edge detection algorithm, and determine the starting point of the target glyph outline; Specifically, in this embodiment, the edge detection algorithm is Canny edge detection, and the starting point is the position where the first stroke of the glyph outline begins; The starting point is overlapped with the starting point of the standard outline of the Chinese character, and a rectangular coordinate system is established with the point as the origin Oe, wherein the standard outline is the outline of the standard writing method of the Chinese character; AS2: Then obtain the coordinates of the turning point, intersection point and end point of the strokes of the standard outline of the Chinese character and the coordinates of the turning point, intersection point and end point of the strokes of the target glyph outline; The turning points and intersection points in the glyph are detected by using the Harris corner point detection method, and the end point is the end position of the last stroke of the glyph outline; AS3: Calculate the deviation value Li between the target glyph outline and the standard glyph outline through the inflection point, intersection point and end point of the matched strokes: Where N is the number of feature points, that is, the total number of turning points and intersection points in the glyph detected by Harris corner point detection plus the end points; ( )and( ) are the feature point coordinates of the target glyph outline and the standard glyph outline respectively; AS4: Then take the average value of the deviation values ​​Li of the Q2 analyzed grids as the standard writing font deviation value Lp of the writing paper this time. The larger the font deviation value Lp, the worse the writing font is. The smaller the font deviation value Lp, the better the writing font is. The calligraphy practice evaluation module is used to establish a calligraphy practice evaluation model based on three core indicators: the error rate, the standard writing deviation coefficient of the current writing paper, and the standard writing font deviation value of the current writing paper. By comparing the calligraphy practice evaluation results of the current writing paper with the historical evaluation results, users can determine whether they have made progress and what needs to be corrected; The calligraphy practice evaluation model is embodied by the following model formula: Among them, S total is the model result score, w1, w2 and w3 are the weight coefficients of the error rate, standard writing deviation coefficient and character shape deviation value respectively. The specific parameters are set by professionals and satisfy w1+w2+w3=1; The calligraphy practice evaluation model is constructed by using three core indicators: the error rate, the standard writing deviation coefficient, and the glyph deviation value. This systematic evaluation method can provide users with a comprehensive evaluation result, so that they can not only know where they are wrong in the practice process, but also understand the overall quality of their writing. In addition, the evaluation model can quantify the user's calligraphy practice effect and compare the current practice results with the historical results to help users intuitively understand their progress or regression. This method of feedback on practice effects through quantitative data can help users see their progress in the long-term practice process and form positive incentives, so as to persist in practice and improve the overall calligraphy level. By using the model result value S for each practice total Compare with the historical model result score value, so that users can quickly understand the effect of this calligraphy practice; Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0019] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. 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 method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A calligraphy practice auxiliary system based on image analysis, characterized in that: include: An image acquisition module, used for acquiring images of the writer practicing on the handwriting paper through a camera; An image preprocessing module is used to preprocess the image in each grid of the acquired screen image to obtain a processed target image, and send it to a subsequent image deviation analysis module; The image deviation analysis module is used to analyze the target image, extract each grid of the target image, and compare it with the corresponding standard grid to determine the writing deviation coefficient of each analyzed grid, and finally obtain the standard writing deviation coefficient of this writing paper, and identify the wrong word grids at the same time, and calculate the wrong word rate of the writing paper according to the total number of grids of the writing paper; The image glyph analysis module uses an edge detection algorithm to extract the glyph outline of the Chinese character in the analyzed grid, determines the starting point of the glyph outline, overlaps it with the starting point of the standard outline by establishing a rectangular coordinate system, and identifies key feature points, and determines the standard writing glyph deviation value of the writing paper by calculation; The calligraphy practice evaluation module is used to establish a calligraphy practice evaluation model based on three core indicators: the error rate, the standard writing deviation coefficient of this writing paper, and the standard writing font deviation value of this writing paper. By comparing the calligraphy practice evaluation results of this writing paper with the historical evaluation results, it is determined whether there is progress and areas that need correction.

2. The calligraphy practice auxiliary system based on image analysis according to claim 1, characterized in that: In the image acquisition module, the camera shoots in the following manner: by placing a handwriting paper of a fixed size at a reference position, the camera shoots it and acquires the image of the practiced calligraphy.

3. The calligraphy practice auxiliary system based on image analysis according to claim 2, characterized in that: The reference position is determined as follows: A two-dimensional rectangular coordinate system is established in the shooting area, and the center of the camera's field of view is used as the origin O of the coordinate system. The horizontal direction of the field of view is the x-axis, and the vertical direction is the y-axis. The field of view of the camera is a rectangular area with a width of W and a length of H. Set the size of the fixed-size handwriting paper to A×B, and there is an m×n boundary grid on the handwriting paper; Align the center of the handwriting paper with the center of the camera's field of view to get the lower left corner P of the handwriting paper. left Coordinates: Then in P left Mark the coordinate point as the reference position; By setting the camera's field of view larger than the size of the handwriting paper: W ≥ A, H ≥ B; If the camera's field of view cannot completely cover the handwriting paper, you need to adjust the distance D between the camera and the handwriting paper to increase the field of view. The camera's field of view is proportional to the distance D: Among them, θ x and θ y are the horizontal and vertical viewing angles of the camera, respectively; Adjust D to ensure that the field of view covers the writing paper.

4. The calligraphy practice auxiliary system based on image analysis according to claim 1, characterized in that: In the image preprocessing module, the preprocessing methods include: Grayscale processing is performed on the image in each grid of the acquired screen image to convert the color image into a grayscale image; Then, the grayscale image in each grid is binarized by setting the threshold using Otsu's method.

5. The calligraphy practice auxiliary system based on image analysis according to claim 1, characterized in that: In the image deviation analysis module, the specific analysis method is as follows: S1: Determine the total number of pixels z in the grid; S2: Select m×n grids on the target image from left to right and from top to bottom as the grids to be analyzed; S3: Identify the Chinese character in the analyzed grid, and find the standard grid corresponding to the Chinese character from the database, wherein the standard grid contains the standard writing of the Chinese character; The specific method of recognizing the Chinese characters in the analyzed grid is as follows: recognizing the Chinese characters in the analyzed grid by optical character recognition technology, marking the Chinese characters as wrong characters if they cannot be recognized; then recognizing the Chinese characters in the next analyzed grid; S4: Compare and analyze the Chinese characters in the analyzed grid with the standard grid to determine the writing deviation coefficient Ci of the analyzed grid; S5: repeating steps S3-S4, determining the number Q1 of specific wrong character grids in the m×n grids, and the writing deviation coefficient Ci of each analyzed grid in the non-wrong character grids; Determine the error rate F by the formula F=Q1 / (m×n); At the same time, determine the number of writing deviation coefficients Ci of the analyzed grids in the non-mistaken character grids as Q2, and Q2=(m×n)-Q1; take the average value of the writing deviation coefficients Ci of Q2 analyzed grids in the non-mistaken character grids as the standard writing deviation coefficient Cp of this writing paper.

6. The calligraphy practice auxiliary system based on image analysis according to claim 5, characterized in that: In step S4, the specific method of determining the writing deviation coefficient Ci of the analyzed grid is: S41: First determine the number of pixels H1 occupied by Chinese characters in the standard grid, and mark the pixels in the grid as Mb i , mark the pixels in the analyzed grid as Mc i ; where 1≤i≤z; S42: Then determine the number k of pixels with the value of 0 that are repeated in the Chinese characters in the standard grid and the analyzed grid, and determine the number j of pixels that do not match, wherein the pixels with the value of 0 in the analyzed grid corresponding to the pixels with the value of 255 in the standard grid are the pixels that do not match; S43: Determine the number of pixels k whose repeated values ​​are 0, and calculate r1=k / H1, where r1 is the boundary grid writing matching rate; Determine the number of unmatched pixels j, and calculate r2=j / H1, where r2 is the boundary grid writing misalignment rate; S44: The writing deviation coefficient Ci of the analyzed grid is calculated according to r1 and r2 by the following formula:

7. The calligraphy practice auxiliary system based on image analysis according to claim 1, characterized in that: In the image glyph analysis module, the standard writing glyph deviation value of the writing paper is determined as follows: AS1: extract the target glyph outline of the Chinese character in the analyzed boundary by using an edge detection algorithm, and determine the starting point of the target glyph outline; The starting point is the position where the first stroke of the glyph outline begins; The starting point is overlapped with the starting point of the standard outline of the Chinese character, and a rectangular coordinate system is established with the point as the origin Oe, wherein the standard outline is the outline of the standard writing method of the Chinese character; AS2: Then obtain the coordinates of the turning point, intersection point and end point of the strokes of the standard outline of the Chinese character and the coordinates of the turning point, intersection point and end point of the strokes of the target glyph outline; The turning points and intersection points in the glyph are detected by using the Harris corner point detection method, and the end point is the end position of the last stroke of the glyph outline; AS3: Calculate the deviation value Li between the target glyph outline and the standard glyph outline through the inflection point, intersection point and end point of the matched strokes: Where N is the number of feature points, which is the total number of turning points and intersection points in the glyph detected by Harris corner detection plus the end points; (x v ,y v ) and (x v ′,y v ′) are the feature point coordinates of the target font outline and the standard font outline respectively; AS4: Then take the average value of the deviation values ​​Li of Q2 analyzed grids as the standard writing font deviation value Lp of this writing paper.

8. The calligraphy practice auxiliary system based on image analysis according to claim 1, characterized in that: In the calligraphy practice evaluation module, the calligraphy practice evaluation model is embodied by the following model formula: S total =100-(w1·F+w2·C p +w3·Lp Among them, S total is the model result score, w1, w2 and w3 are the weight coefficients of the error rate, standard writing deviation coefficient and character shape deviation value respectively, and they satisfy w1+w2+w3=1.

Images