Automatic color matching method for children's electronic drawing board

Abstract

The present invention relates to the field of automatic color matching technology, and in particular, to an automatic color matching method for a children's electronic drawing board. The method comprises the following steps: obtaining a drawing graphic template and drawing color matching history data selected by a child to calculate the curvature of the graphic, using the calculated graphic curvature and a set graphic curvature threshold to judge the discontinuity of the graphic; and calculating the similarity through the coordinates of the data points in the child's drawing graphic and the coordinates of the data points in the selected drawing graphic template. The present invention extracts style features based on the analyzed drawing style data, and then inputs the extracted style feature data, the extracted key feature vector data and the similarity into a neural network model to predict the coordinate point set of the graphic, and generates the graphic that the child wants to draw based on the predicted graphic coordinate point set. By generating the graphic that the child wants to draw through the predicted graphic coordinate point set, the manual adjustment process can be omitted, and the automatic color matching and filling become smoother, thereby improving efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic color matching, and in particular to an automatic color matching method for a children's electronic drawing board. Background Art

[0002] The automatic color matching of children's electronic drawing board can integrate a variety of technologies and methods to realize intelligent and automatic color matching and filling functions. It can realize intelligent color filling and personalized experience by combining image processing technology, machine learning, intelligent recommendation and user interaction functions, and help them learn and master color usage skills. When children draw graphics on the electronic board according to the graphics provided by the drawing template, due to the immature hand coordination of children, there are discontinuous lines in the drawing process, which makes the boundaries of the drawing graphics incomplete and unable to accurately identify the boundaries of the graphics, resulting in missing or incorrect filling when the internal area of ​​the graphics is filled with colors, thereby reducing the accuracy of color filling. In order to avoid the reduction of color filling accuracy caused by discontinuous lines in the drawing process, we provide an automatic color matching method for children's electronic drawing board. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatic color matching of a children's electronic drawing board to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides a method for automatically matching colors of a children's electronic drawing board, comprising the following steps:

[0005] S1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching and drawing style, establish a drawing model based on the analyzed preferred color matching and drawing style data and drawing color matching history data, and then obtain the data point coordinates and corresponding timestamps in the graphic data of the child's drawing through electromagnetic induction, calculate the graphic curvature, and use the calculated graphic curvature and the set graphic curvature threshold to determine the graphic discontinuity;

[0006] S2, triggering a feature extraction formula through the presence of intermittent data in the drawing, extracting key feature vectors based on the data point coordinates, corresponding timestamps, and calculated graphic curvature, and then calculating similarity between the data point coordinates in the children's drawing and the data point coordinates in the selected drawing graphic template;

[0007] S3. Use a clustering algorithm to extract style features based on the analyzed painting style data, and then predict the graphics that the child wants to draw through the style feature data extracted by the neural network model, the extracted key feature vector data and the similarity. Finally, automatically match the color of the predicted painting graphics according to the color matching of the graphics in the painting graphic template.

[0008] As a further improvement of the present technical solution, the S1 specifically includes the following method steps:

[0009] S1.1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching in the past week based on the drawing color matching history data, and analyze the drawing style in the past week, and use a convolutional neural network to establish a drawing model based on the analyzed preferred color matching data, the analyzed drawing style and the drawing history data, so as to store the preferred color matching of the child for various graphics in the past week, and then automatically match the preferred color to the selected drawing graphic template based on the drawing model, and fill the color matching into the drawing graphic template;

[0010] S1.2. Obtain the graphic data of the children's drawings, the data point coordinates of the graphic line segments and the pressure data applied by the electronic pen through electromagnetic induction, and calculate the graphic curvature based on the graphic data of the children's drawings using the curvature algorithm formula.

[0011] As a further improvement of the technical solution, the implementation steps of calculating the curvature in S1.2 of S1 are as follows:

[0012] Step 1: Collect graphic data of children’s paintings and the data point coordinates of the graphic segment , and uses electromagnetic induction technology to read the graphic data of children's paintings Get the coordinates of the data point corresponding to the graphic line segment Coordinates of adjacent data points , and the data point coordinates of the graphic segment The coordinates of the adjacent data points The corresponding timestamp ;

[0013] Step ②: According to the data point coordinates of the graphic line segment The coordinates of the adjacent data points Subtract the corresponding timestamps to get the time interval , and then use the data point coordinates of the graphic line segment , adjacent data point coordinates and time interval Calculate the derivative of the graphics coordinates and get the derivative of the graphics coordinates , the specific algorithm formula is:

[0014] ;

[0015] ;

[0016] Among them, i refers to the i-th data point in the graph segment;

[0017] Step 3: Use the graphic coordinate derivative and time interval Calculate the second-order derivative of the graphic coordinates and obtain the second-order derivative of the graphic coordinates , the specific algorithm formula is:

[0018] ;

[0019] ;

[0020] Step 4: Use the graphic coordinate derivative and the second derivative of the graphic coordinates Calculate the curvature of the figure and obtain the calculated curvature of the figure , the specific algorithm formula is:

[0021] ;

[0022] in, Refers to the calculation of The formula is used to calculate the curvature of a graphic. The curvature of the graphic can be used to determine whether the graphic is discontinuous. Automatic color matching and filling greatly reduce the time for manual drawing and coloring. Children can see their drawing results faster, thereby increasing their interest and participation.

[0023] As a further improvement of the present technical solution, S1.3, the discontinuity of the graphic is judged based on the calculated graphic curvature and the set graphic curvature threshold. When the calculated graphic curvature is greater than the set graphic curvature threshold, it indicates that the painted graphic is discontinuous. When the calculated graphic curvature is less than the set graphic curvature threshold, it indicates that the painted graphic is not discontinuous. The painted graphic is automatically colored according to the painting graphic template, and the color is filled into the painted graphic.

[0024] As a further improvement of the present technical solution, the S2 specifically includes the following method steps:

[0025] S2.1. The intermittent data in the drawing triggers a feature extraction formula, and the feature extraction formula is used to extract key feature vectors according to the data point coordinates of the graphic line segment and the corresponding timestamp, the pressure data applied by the electronic pen, and the calculated graphic curvature;

[0026] S2.2. Obtain the coordinates of the data points in the children's drawing and the selected drawing graphic template through electromagnetic induction technology, and then calculate the variance using the coordinates of the data points in the children's drawing and the coordinates of the data points in the selected drawing graphic template. Then calculate the similarity based on the coordinates of the data points in the children's drawing, the coordinates of the data points in the drawing graphic template and the variance, and calculate the similarity between the children's drawing and the drawing in the drawing graphic template.

[0027] As a further improvement of the technical solution, the implementation principle of calculating the similarity in S2.2 of S2 is:

[0028] Get the coordinates of data points in children's drawing graphics , drawing graphic template The coordinates of the data points in and variance Calculate similarity and get similarity , the specific algorithm formula is:

[0029] ;

[0030] in, Refers to the children's drawing graphics and drawing graphic templates The coordinates of the data points, Refers to the horizontal axis coordinate of the i-th data point in the drawing graphic template. It refers to the vertical coordinate of the coordinates of the ith data point in the drawing graphic template. This formula is used to calculate the similarity between the graphics drawn by children and the graphics in the drawing graphic template. By calculating the similarity, instant feedback can be provided to increase interactivity. Children can quickly understand the degree of similarity between their own paintings and the template, and make adjustments and improvements accordingly.

[0031] As a further improvement of the present technical solution, S3 specifically includes the following method steps:

[0032] S3.1. Using a clustering algorithm to extract style features from the analyzed painting style data, and then inputting the extracted style feature data, the extracted key feature vector data and the similarity into a neural network model. The neural network model learns the extracted style feature data, the extracted key feature vector data and the similarity, and predicts the coordinate point set of the output graphic according to the learning result ("coordinate point set" refers to a set of coordinate data representing each data point in a graphic or image), and generates the graphic that the child wants to draw according to the predicted graphic coordinate point set;

[0033] S3.2. Automatically color matching the generated drawing graphic according to the graphic color matching in the drawing graphic template, and fill the color matching into the generated drawing graphic.

[0034] As a further improvement of the technical solution, the realization principle of the coordinate point set of the predicted output graphics in S3.1 of S3 is:

[0035] Collect the extracted style feature data , extracted key feature vector data and similarity , and through the neural network model Predict the coordinate point set of the output graphics and obtain the predicted graphics coordinate point set , the specific algorithm formula is:

[0036] ;

[0037] Among them, the formula is used to predict the graphic coordinate point set. By predicting the coordinate point set through the model, the complete graphic data can be quickly generated, thereby speeding up the automatic color matching and filling, allowing children to see their painting results faster.

[0038] The second object of the present invention is to provide a system for operating any of the above-mentioned automatic color matching methods for children's electronic drawing boards, including a modeling calculation unit, a similarity extraction unit, and a data prediction unit;

[0039] The modeling calculation unit is used to obtain the drawing graphic template and drawing color matching history selected by the child to establish a drawing model, and then automatically match the preferred color according to the drawing model and fill it into the selected drawing graphic template, obtain the graphic data of the child's drawing through electromagnetic induction and calculate the graphic curvature, and determine whether the drawing graphic is discontinuous according to the calculated graphic curvature;

[0040] The similarity extraction unit is used to receive the intermittent data of the graphics drawn in the modeling calculation unit to trigger the feature extraction formula, extract the key feature vector, and then obtain the data point coordinates in the children's drawing graphics and the data point coordinates in the selected drawing graphics template through electromagnetic induction technology to calculate the variance, and calculate the graphics similarity according to the data point coordinates in the children's drawing graphics, the data point coordinates in the drawing graphics template and the variance;

[0041] The data prediction unit is used to receive the key feature vector data and similarity extracted in the similarity extraction unit, and input them into the neural network model. The neural network model predicts and outputs the graphics that the child wants to draw by learning the extracted key feature vector data and similarity, and then automatically colors and fills the drawing graphics accordingly.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. The automatic color matching method of the children's electronic drawing board obtains the painting graphic template and painting color matching history data selected by the child, analyzes the preferred color matching and painting style, establishes a painting model based on the analyzed preferred color matching and painting style data and the painting color matching history data, and then calculates the graphic curvature through the data point coordinates and the corresponding timestamps in the graphic data of the children's painting, and uses the calculated graphic curvature and the set graphic curvature threshold to judge the discontinuity of the graphic. By calculating the precise curvature and judging the discontinuity of the graphic, the color matching can be accurately filled into the area inside the graphic, avoiding missing or incorrect filling of the internal area of ​​the graphic according to the preferred color matching, ensuring that the internal area of ​​the graphic is completely filled, and improving the integrity of color matching and precise filling of the graphic.

[0044] 2. The automatic color matching method of the children's electronic drawing board extracts style features based on the analyzed painting style data, and then inputs the extracted style feature data, the extracted key feature vector data and the similarity into the neural network model to predict the coordinate point set of the graphic, and generates the graphic that the child wants to draw based on the predicted graphic coordinate point set. By generating the graphic that the child wants to draw by predicting the coordinate point set of the graphic, the manual adjustment process can be omitted, and the automatic color matching and filling become smoother, thereby improving efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flowchart of the overall steps of the present invention;

[0046] Figure 2 It is an overall block diagram of the present invention.

[0047] The meaning of each number in the figure is:

[0048] 1. Modeling and calculation unit; 2. Extracting similarity unit; 3. Data prediction unit. DETAILED DESCRIPTION

[0049] The following will be combined with the accompanying drawings in 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.

[0050] Embodiment 1: The present invention provides a method for automatically matching colors on a children's electronic drawing board. Figure 1 , comprising the following method steps:

[0051] S1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching and drawing style, establish a drawing model based on the analyzed preferred color matching and drawing style data and drawing color matching history data, and then obtain the data point coordinates and corresponding timestamps in the graphic data of the child's drawing through electromagnetic induction, calculate the graphic curvature, and use the calculated graphic curvature and the set graphic curvature threshold to determine the graphic discontinuity;

[0052] S1.1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching in the past week based on the drawing color matching history data, and analyze the drawing style in the past week, and use a convolutional neural network to establish a drawing model based on the analyzed preferred color matching data, the analyzed drawing style, and the drawing history data (including the historical color matching data and the historical drawing style), so as to store the preferred color matching of the child for various graphics in the past week, and then automatically match the preferred color to the selected drawing graphic template based on the drawing model, and fill the color matching into the drawing graphic template;

[0053] S1.2, obtaining the graphic data of the children's drawings, the data point coordinates of the graphic segments and the pressure data applied by the electronic pen through electromagnetic induction, and calculating the graphic curvature based on the graphic data of the children's drawings using the curvature algorithm formula, and calculating the similarity through the coordinate points, so as to accurately determine the similarity between the children's drawing graphics and the template graphics, which helps to more accurately identify the theme or style of the painting and match it with the predefined template;

[0054] Implementation steps for calculating curvature:

[0055] Step 1: Collect graphic data of children’s paintings and the data point coordinates of the graphic segment , and uses electromagnetic induction technology to read the graphic data of children's paintings Get the coordinates of the data point corresponding to the graphic line segment Coordinates of adjacent data points , and the data point coordinates of the graphic segment The coordinates of the adjacent data points The corresponding timestamp ;

[0056] Step ②: According to the data point coordinates of the graphic line segment The coordinates of the adjacent data points Subtract the corresponding timestamps to get the time interval , and then use the data point coordinates of the graphic line segment , adjacent data point coordinates and time interval Calculate the derivative of the graphics coordinates and get the derivative of the graphics coordinates , the specific algorithm formula is:

[0057] ;

[0058] ;

[0059] Among them, i refers to the i-th data point in the graph segment;

[0060] Step 3: Use the graphic coordinate derivative and time interval Calculate the second-order derivative of the graphic coordinates and obtain the second-order derivative of the graphic coordinates , the specific algorithm formula is:

[0061] ;

[0062] ;

[0063] Step 4: Use the graphic coordinate derivative and the second derivative of the graphic coordinates Calculate the curvature of the figure and obtain the calculated curvature of the figure , the specific algorithm formula is:

[0064] ;

[0065] in, Refers to the calculation of The formula is used to calculate the curvature of a drawing figure. The curvature of the figure can be used to determine whether the figure is discontinuous. Automatic color matching and filling greatly reduce the time of manual drawing and coloring. Children can see their own drawing results faster, thereby increasing their interest and participation.

[0066] S1.3. The discontinuity of the graphic is judged according to the calculated graphic curvature and the set graphic curvature threshold. When the calculated graphic curvature is greater than the set graphic curvature threshold, it indicates that the graphic being drawn is discontinuous. When the calculated graphic curvature is less than the set graphic curvature threshold, it indicates that the graphic being drawn is not discontinuous. The graphic being drawn is automatically color-matched according to the drawing graphic template, and the color-matched is filled into the drawing graphic.

[0067] S2, triggering a feature extraction formula through the presence of intermittent data in the drawing, extracting key feature vectors based on the data point coordinates, corresponding timestamps, and calculated graphic curvature, and then calculating similarity between the data point coordinates in the children's drawing and the data point coordinates in the selected drawing graphic template;

[0068] S2.1. The intermittent data in the drawing triggers the feature extraction formula. The key feature vectors are extracted based on the data point coordinates of the line segments and the corresponding timestamps, the pressure data applied by the electronic pen, and the calculated curvature of the drawing. The extracted feature vectors can help detect whether the drawing is complete, especially the interruptions or omissions that occur during the children's drawing process, so as to ensure that the generated drawing is complete;

[0069] The implementation principle of extracting key features using feature extraction formula:

[0070] First, collect the data point coordinates of the graphic line segment and the corresponding timestamp , pressure data applied by the electronic pen , calculated figure curvature Extract key features and obtain the extracted key feature vector data , the specific algorithm formula is:

[0071] ;

[0072] Among them, the graphic data of children's paintings Contains n data point coordinates. The formula is used to extract key feature vectors. Through the extraction of key feature vectors, it is used to automate the generation and optimization of graphics, thereby improving efficiency and consistency.

[0073] S2.2, obtaining the coordinates of the data points in the children's drawing and the coordinates of the data points in the selected drawing graphic template through electromagnetic induction technology, and then calculating the variance using the coordinates of the data points in the children's drawing and the coordinates of the data points in the selected drawing graphic template, and then calculating the similarity based on the coordinates of the data points in the children's drawing, the coordinates of the data points in the drawing graphic template and the variance, and calculating the similarity between the children's drawing and the drawing in the drawing graphic template. Based on the similarity results, personalized feedback and suggestions can be provided, for example, if the children's drawing is similar to the drawing in the template, it is suggested to add certain details or try new techniques to further improve the drawing effect;

[0074] The implementation principle of calculating similarity:

[0075] Get the coordinates of data points in children's drawing graphics , drawing graphic template The coordinates of the data points in and variance Calculate similarity and get similarity , the specific algorithm formula is:

[0076] ;

[0077] in, Refers to the children's drawing graphics and drawing graphic templates The coordinates of the data points, Refers to the horizontal axis coordinate of the i-th data point in the drawing graphic template. Refers to the vertical coordinate of the coordinates of the i-th data point in the drawing graphic template. This formula is used to calculate the similarity between the graphics drawn by children and the graphics in the drawing graphic template. By calculating the similarity, feedback can be provided instantly to increase interactivity. Children can quickly understand the similarity between their own drawings and the template, and thus make adjustments and improvements.

[0078] S3, using a clustering algorithm to extract style features based on the analyzed painting style data, and then predicting the graphics that the child wants to draw through the style feature data extracted by the neural network model, the extracted key feature vector data and the similarity, and finally automatically matching the predicted painting graphics according to the color matching of the graphics in the painting graphic template;

[0079] S3.1. Use a clustering algorithm to extract style features based on the analyzed painting style data, and then input the extracted style feature data, extracted key feature vector data and similarity into the neural network model. The neural network model predicts the coordinate point set of the output graphic based on the learning result by learning the extracted style feature data, extracted key feature vector data and similarity ("coordinate point set" refers to a set of coordinate data representing each data point in a graphic or image), and generates the graphic that the child wants to draw based on the predicted graphic coordinate point set. The coordinate point set of the graphic is accurately predicted based on the style features and key feature vectors, which means that the generated graphic will be more in line with the child's creativity and design intention, thereby improving the accuracy of automatic color matching and filling;

[0080] The realization principle of the coordinate point set of the predicted output graphics:

[0081] Collect the extracted style feature data , extracted key feature vector data and similarity , and through the neural network model Predict the coordinate point set of the output graphics and obtain the predicted graphics coordinate point set , the specific algorithm formula is:

[0082] ;

[0083] Among them, the formula is used to predict the coordinate point set of the graphic. By predicting the coordinate point set through the model, the complete graphic data can be quickly generated, thereby speeding up the automatic color matching and filling, so that children can see their painting results faster;

[0084] S3.2. Automatically color matching the generated drawing graphic according to the graphic color matching in the drawing graphic template, and fill the color matching into the generated drawing graphic.

[0085] The second object of the present invention is to provide a system for operating the above-mentioned automatic color matching method of the children's electronic drawing board, comprising a modeling calculation unit 1, a similarity extraction unit 2, and a data prediction unit 3;

[0086] The following is a description of the modeling calculation unit 1, the similarity extraction unit 2 and the data prediction unit 3. Figure 2 ;

[0087] The modeling calculation unit 1 is used to obtain the drawing graphic template and the drawing color matching history selected by the child to establish a drawing model, and then automatically match the preferred color according to the drawing model and fill it into the selected drawing graphic template, obtain the graphic data of the child's drawing through electromagnetic induction and calculate the graphic curvature, and judge whether the drawing graphic is discontinuous according to the calculated graphic curvature;

[0088] The similarity extraction unit 2 is used to receive the intermittent data of the drawing in the modeling calculation unit 1 to trigger the feature extraction formula, extract the key feature vector, and then obtain the data point coordinates in the children's drawing and the data point coordinates in the selected drawing graphic template through electromagnetic induction technology to calculate the variance, and calculate the graphic similarity according to the data point coordinates in the children's drawing, the data point coordinates in the drawing graphic template and the variance;

[0089] The data prediction unit 3 is used to receive the key feature vector data and similarity extracted in the similarity extraction unit 2, and input them into the neural network model. The neural network model predicts and outputs the graphics that the child wants to draw by learning the extracted key feature vector data and similarity, and then automatically colors and fills the drawing graphics accordingly.

[0090] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for automatic color matching of a children's electronic drawing board, characterized in that: The method comprises the following steps: S1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching and drawing style, establish a drawing model based on the analyzed preferred color matching and drawing style data and drawing color matching history data, and then obtain the data point coordinates and corresponding timestamps in the graphic data of the child's drawing through electromagnetic induction, calculate the graphic curvature, and use the calculated graphic curvature and the set graphic curvature threshold to determine the graphic discontinuity; S1.

1. Obtain the drawing graphic template and drawing color matching history data selected by the child, analyze the preferred color matching in the past week based on the drawing color matching history data, and analyze the drawing style in the past week, and use a convolutional neural network to establish a drawing model based on the analyzed preferred color matching data, the analyzed drawing style and the drawing history data, so as to store the preferred color matching of the child for various graphics in the past week, and then automatically match the preferred color to the selected drawing graphic template based on the drawing model, and fill the color matching into the drawing graphic template; S1.2, obtaining the graphic data of the children's drawing, the data point coordinates of the graphic line segments and the pressure data applied by the electronic pen through electromagnetic induction, and calculating the graphic curvature according to the graphic data of the children's drawing using a curvature algorithm formula; Implementation steps for calculating curvature: Step 1: Collect graphic data of children’s paintings and the data point coordinates of the graphic segment , and uses electromagnetic induction technology to read the graphic data of children's paintings Get the coordinates of the data point corresponding to the graphic line segment Coordinates of adjacent data points , and the data point coordinates of the graphic segment The coordinates of the adjacent data points The corresponding timestamp ; Step ②: According to the data point coordinates of the graphic line segment The coordinates of the adjacent data points Subtract the corresponding timestamps to get the time interval , and then use the data point coordinates of the graphic line segment , adjacent data point coordinates and time interval Calculate the derivative of the graphics coordinates and get the derivative of the graphics coordinates , the specific algorithm formula is: ； ； Among them, i refers to the i-th data point in the graph segment; Step 3: Use the graphic coordinate derivative and time interval Calculate the second-order derivative of the graphic coordinates and obtain the second-order derivative of the graphic coordinates , the specific algorithm formula is: ； ； Step 4: Use the graphic coordinate derivative and the second derivative of the graphic coordinates Calculate the curvature of the figure and obtain the calculated curvature of the figure , the specific algorithm formula is: ； in, Refers to the calculation of The curvature of the graph; S1.

3. The discontinuity of the graphic is judged according to the calculated graphic curvature and the set graphic curvature threshold. When the calculated graphic curvature is greater than the set graphic curvature threshold, it indicates that the graphic being drawn is discontinuous. When the calculated graphic curvature is less than the set graphic curvature threshold, it indicates that the graphic being drawn is not discontinuous. The graphic being drawn is automatically color-matched according to the drawing graphic template, and the color-matched is filled into the drawing graphic. S2, triggering a feature extraction formula through the presence of intermittent data in the drawing, extracting key feature vectors based on the data point coordinates, corresponding timestamps, and calculated graphic curvature, and then calculating similarity between the data point coordinates in the children's drawing and the data point coordinates in the selected drawing graphic template; S2.

1. The intermittent data in the drawing triggers a feature extraction formula, and the feature extraction formula is used to extract key feature vectors according to the data point coordinates of the graphic line segment and the corresponding timestamp, the pressure data applied by the electronic pen, and the calculated graphic curvature; The implementation principle of extracting key features using feature extraction formula: First collect the data point coordinates of the graphic segment and the corresponding timestamp , pressure data applied by the electronic pen , calculated graph curvature Extract key features and obtain extracted key feature vector data , the specific algorithm formula is: ； Among them, the graphic data of children's paintings Contains the coordinates of n data points; S3. Use a clustering algorithm to extract style features based on the analyzed painting style data, and then predict the graphics that the child wants to draw through the style feature data extracted by the neural network model, the extracted key feature vector data and the similarity. Finally, automatically match the color of the predicted painting graphics according to the color matching of the graphics in the painting graphic template.

2. The automatic color matching method for children's electronic drawing board according to claim 1, characterized in that: The S2 specifically includes the following method steps: S2.

2. Obtain the coordinates of the data points in the children's drawing and the selected drawing graphic template through electromagnetic induction technology, and then calculate the variance using the coordinates of the data points in the children's drawing and the coordinates of the data points in the selected drawing graphic template. Then calculate the similarity based on the coordinates of the data points in the children's drawing, the coordinates of the data points in the drawing graphic template and the variance, and calculate the similarity between the children's drawing and the drawing in the drawing graphic template.

3. The automatic color matching method for children's electronic drawing board according to claim 2, characterized in that: The implementation principle of calculating similarity in S2.2 of S2 is: Get the coordinates of data points in children's drawing graphics , drawing graphic template The coordinates of the data points in and variance Calculate similarity and get similarity , the specific algorithm formula is: ； in, Refers to the children's drawing graphics and drawing graphic templates The coordinates of the data points, Refers to the horizontal axis coordinate of the i-th data point in the drawing graphic template. Refers to the vertical coordinate of the i-th data point in the drawing graphic template.

4. The automatic color matching method for children's electronic drawing board according to claim 1, characterized in that: The S3 specifically includes the following method steps: S3.

1. Using a clustering algorithm to extract style features based on the analyzed painting style data, and then inputting the extracted style feature data, the extracted key feature vector data and the similarity into a neural network model. The neural network model learns the extracted style feature data, the extracted key feature vector data and the similarity, predicts a coordinate point set of an output graphic based on the learning result, and generates a graphic that the child wants to draw based on the predicted graphic coordinate point set; S3.

2. Automatically color matching the generated drawing graphic according to the graphic color matching in the drawing graphic template, and fill the color matching into the generated drawing graphic.

5. The automatic color matching method for children's electronic drawing board according to claim 4, characterized in that: The realization principle of the coordinate point set of the predicted output graphics in S3.1 of S3 is as follows: Collect the extracted style feature data , extracted key feature vector data and similarity , and through the neural network model Predict the coordinate point set of the output graphics and obtain the predicted graphics coordinate point set , the specific algorithm formula is: .

6. A system for operating the automatic color matching method for children's electronic drawing board according to any one of claims 1 to 5, characterized in that: It includes a modeling calculation unit (1), a similarity extraction unit (2), and a data prediction unit (3); The modeling calculation unit (1) is used to obtain the drawing graphic template and drawing color matching history selected by the child to establish a drawing model, then automatically match the preferred color according to the drawing model and fill it into the selected drawing graphic template, obtain the graphic data of the child's drawing through electromagnetic induction and calculate the graphic curvature, and determine whether the drawing graphic is discontinuous according to the calculated graphic curvature; The similarity extraction unit (2) is used to receive the intermittent data of the graphics drawn in the modeling calculation unit (1) to trigger the feature extraction formula, extract the key feature vector, and then obtain the data point coordinates in the children's drawing graphics and the data point coordinates in the selected drawing graphics template through electromagnetic induction technology to calculate the variance, and calculate the graphics similarity based on the data point coordinates in the children's drawing graphics, the data point coordinates in the drawing graphics template and the variance; The data prediction unit (3) is used to receive the key feature vector data and similarity extracted in the similarity extraction unit (2) and input them into the neural network model. The neural network model predicts and outputs the figure that the child wants to draw by learning the extracted key feature vector data and similarity, and then automatically matches the color of the drawing figure and fills it into the drawing figure accordingly.

Images