A method and system for improving the contour matching degree of a machine vision image

Abstract

The application provides a method and system for improving the matching degree of machine vision image contours, applied to the technical field of intelligent control, and the method comprises the following steps: acquiring a target image by connecting an image acquisition device terminal; performing image edge extraction on the target image to obtain image edge features; inputting the image edge features into a cloud processor for analysis to obtain a contour extraction result; inputting the contour extraction result into a gray scale conversion module to perform gray scale conversion and output a gray scale conversion image; performing gray scale layer analysis on the gray scale conversion image to determine contour layer pixel coordinates; outputting a matching contour layer of the target image based on the contour layer pixel coordinates to realize contour matching; the image contour extraction precision is improved, and the matching accuracy of the target image contour is improved; the technical problem of low image contour extraction precision in the prior art, which leads to low accuracy of object recognition through contours, is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital image processing, and particularly relates to a method and system for improving contour matching degree of machine vision image. BACKGROUND

[0002] Machine vision is to convert a target into an image signal through an image acquisition device, to obtain information features contained in the target through various operations on the image signal by an image processing system, and then to make corresponding processing according to the information features. However, in the prior art, the contour is directly extracted by using an edge extraction method, and the object contour is directly obtained through edge coordinates, so that the contour extraction precision is low, and the object recognition through the contour is prone to error and the recognition precision is low.

[0003] Therefore, in the prior art, there is a technical problem of low image contour extraction precision, which leads to low accuracy of object recognition through the contour. SUMMARY

[0004] The present application provides a method and system for improving contour matching degree of machine vision image, which is used to solve the technical problem of low image contour extraction precision in the prior art, which leads to low accuracy of object recognition through the contour.

[0005] In view of the above problems, the present application provides a method and system for improving contour matching degree of machine vision image.

[0006] In a first aspect, the present application provides a method for improving contour matching degree of machine vision image, which is applied to an image recognition intelligent processing system, the system is in communication connection with a cloud processor, and the method comprises the following steps: connecting an image acquisition device terminal to obtain a target image to be processed, wherein the target image is an image that needs to be extracted; performing image edge extraction on the target image to obtain image edge features; inputting the image edge features into the cloud processor, analyzing according to a contour extraction model chained in the cloud processor, and obtaining a contour extraction result output by the contour extraction model; inputting the contour extraction result into a gray scale conversion module, performing gray scale conversion on the contour extraction result based on the gray scale conversion module, and outputting a gray scale conversion image; determining contour layer pixel coordinates by performing gray scale layer analysis on the gray scale conversion image; and outputting a matching contour layer of the target image with the contour layer pixel coordinates, so as to realize contour matching.

[0007] In a second aspect of the present application, a system for improving the contour matching degree of a machine vision image is provided, which is in communication with a cloud processor, and includes: an image acquisition module configured to connect an image acquisition device terminal and acquire a target image to be processed, wherein the target image is an image that needs to be subjected to contour extraction; an image edge feature acquisition module configured to perform image edge extraction on the target image and acquire image edge features; a contour extraction module configured to input the image edge features into the cloud processor, analyze based on a contour extraction model chained in the cloud processor, and acquire a contour extraction result output by the contour extraction model; a grayscale conversion module configured to input the contour extraction result into the grayscale conversion module, perform grayscale conversion on the contour extraction result based on the grayscale conversion module, and output a grayscale conversion image; a contour layer pixel coordinate acquisition module configured to determine contour layer pixel coordinates by performing grayscale layer analysis on the grayscale conversion image; and a contour matching module configured to output a matching contour layer of the target image based on the contour layer pixel coordinates, so as to realize contour matching.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] The method provided in the present application acquires a target image by connecting an image acquisition device terminal. Image edge extraction is performed on the target image to acquire image edge features. The image edge features are input into a cloud processor for analysis to acquire a contour extraction result. The contour extraction result is input into a grayscale conversion module for grayscale conversion to output a grayscale conversion image. Contour layer pixel coordinates are determined by performing grayscale layer analysis on the grayscale conversion image. A matching contour layer of the target image is output based on the contour layer pixel coordinates to realize contour matching. Contour layer pixel coordinates are acquired based on the contour extraction result, which improves the image contour extraction precision and further improves the matching accuracy of the target image contour. The technical problem of low image contour extraction precision in the prior art, which leads to low accuracy of object recognition by contour, is solved.

[0010] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the following detailed description of the embodiments of the present application can be implemented in accordance with the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following detailed description of the embodiments of the present application is provided. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 A method for improving the contour matching degree of a machine vision image is provided.

[0012] Figure 2A schematic diagram of the process for obtaining contour matching results in a method for improving contour matching accuracy of machine vision images provided in this application;

[0013] Figure 3 A schematic diagram of the process for obtaining the pixel coordinates of the contour layer in a method for improving the contour matching degree of machine vision images provided in this application;

[0014] Figure 4 This application provides a schematic diagram of a system structure for improving the contour matching accuracy of machine vision images.

[0015] Figure labeling: Image acquisition module 11, Image edge feature acquisition module 12, Contour extraction module 13, Grayscale conversion module 14, Contour layer pixel coordinate acquisition module 15, Contour matching module 16. Detailed Implementation

[0016] This application provides a method and system for improving the contour matching accuracy of machine vision images, which addresses the technical problem in the prior art where the low precision of image contour extraction leads to a low accuracy rate in object recognition based on contours.

[0017] The technical solutions in this application will now be clearly and completely described with reference to the accompanying drawings. The described embodiments are only a part of what can be achieved by this application, and not all of the contents of this application.

[0018] Example 1

[0019] like Figure 1 As shown, this application provides a method for improving the contour matching accuracy of machine vision images. The method is applied to an image recognition intelligent processing system, which is communicatively connected to a cloud processor. The method includes:

[0020] Step 100: Connect the image acquisition device terminal to acquire the target image to be processed, wherein the target image is the image for which contour extraction is required;

[0021] Step 200: Extract image edges from the target image to obtain image edge features;

[0022] Step 300: Input the image edge features into the cloud processor, analyze them according to the contour extraction model linked into the cloud processor, and obtain the contour extraction result output by the contour extraction model;

[0023] Specifically, an image acquisition device terminal is connected to acquire the target image to be acquired, where the target image is the image for contour extraction in this embodiment. Subsequently, image edge extraction is performed on the acquired target image to obtain image edge features. Common edge extraction algorithms can be used for edge extraction to obtain image edge features. Further, the image edge features are input into the cloud processor, where they are analyzed by a contour extraction model linked to the cloud processor. That is, the contour extraction model in the cloud processor analyzes the image edge features, and the contour results corresponding to the image edge features are obtained, yielding the contour extraction result output by the contour extraction model. In other words, based on the image edge features, the contour in the target image is obtained to obtain the contour extraction result. Since the accuracy of contour extraction based on image edge features is relatively low, the contour in the target image is initially obtained through image edge features, and then further processing is performed on the contour to obtain more accurate contour pixel coordinates.

[0024] like Figure 2 As shown, the method step 100 provided in this application embodiment further includes:

[0025] Step 110: Input the target image into the image quality detection model, and obtain the quality detection result according to the image quality detection model, wherein the quality detection result is a result used to identify the quality level of the target image;

[0026] Step 120: If the quality inspection result fails, obtain a secondary image acquisition command;

[0027] Step 130: According to the image secondary acquisition instruction, replace and cover the target image until the quality detection result is passed.

[0028] Specifically, in the target image input image quality detection model, a quality detection result is obtained based on the model. This quality detection result is used to identify the quality level of the target image. By performing image quality detection on the target image, the processed target image meets predetermined requirements, ensuring the accuracy of contour recognition. If the quality detection result fails, a secondary image acquisition command is obtained, and the image is acquired again. Based on the secondary image acquisition command, the target image is replaced and covered until the quality detection result passes, thereby ensuring the target image meets predetermined requirements and improving the accuracy of contour recognition.

[0029] The method step 130 provided in this application embodiment further includes:

[0030] Step 131: Analyze the target image according to the image quality detection model to obtain image sharpness and color level correlation;

[0031] Step 132: Based on the image sharpness and the color level correlation as dual-value conditions, perform dual-value logic decision analysis and output decision results, wherein the decision results include pass and fail;

[0032] Step 133: Obtain the quality inspection results based on the decision results.

[0033] Specifically, the target image is analyzed using an image quality detection model to obtain its sharpness and color gradation correlation. Since the image is used for contour extraction in this embodiment, image sharpness is equivalent to image resolution; higher resolution results in more clearly defined image details. Color gradation correlation is an index of image brightness; higher color gradations result in more prominent color features. In practical applications, image sharpness and color gradation correlation can be set to fixed thresholds based on actual conditions. Finally, using image sharpness and color gradation correlation as dual-value conditions, a dual-value logic decision-maker is used for analysis. The target image must simultaneously meet the aforementioned adjustments for image sharpness and color gradation correlation, outputting a decision result. This decision result includes pass and fail, and the quality detection result is obtained based on the decision result. This ensures the target image meets predetermined requirements and improves the accuracy of contour recognition.

[0034] Step 400: Input the contour extraction result into the grayscale conversion module, perform grayscale conversion on the contour extraction result based on the grayscale conversion module, and output the grayscale converted image;

[0035] Step 500: Determine the pixel coordinates of the contour layer by performing grayscale layer analysis on the grayscale converted image;

[0036] Step 600: Output the matching contour layer of the target image using the pixel coordinates of the contour layer to achieve contour matching.

[0037] Specifically, the contour extraction results output by the contour extraction model are input into the grayscale conversion module. The grayscale conversion module performs grayscale conversion on the contour extraction results and outputs a grayscale converted image. Subsequently, grayscale layer analysis is performed on the acquired grayscale converted image to determine the pixel coordinates of the contour layer in the image. During grayscale layer analysis, the grayscale data variation characteristics of the grayscale layer are obtained to determine the grayscale value range of the contour edges. Further analysis of the grayscale value range determines the final contour layer pixel coordinates; that is, all pixel coordinates of the contour are obtained based on the contour edge coordinates. Finally, using the contour layer pixel coordinates, the matching contour layer of the target image is output, i.e., the matching contour layer of the target image used for subsequent contour matching. Contour matching of the target image is then achieved through this matching contour layer. By obtaining the contour edges for matching through edge extraction and obtaining the pixel coordinates of the contours based on these edges, the accuracy of image contour extraction is improved, thereby improving the accuracy of contour matching of the target image.

[0038] The method step 600 provided in this application embodiment further includes:

[0039] Step 700: Access the image matching system and obtain a matching image, wherein the matching image is an image used to match the target image;

[0040] Step 710: Analyze the matching image and output the matching contour layer of the matching image;

[0041] Step 720: Perform contour matching between the matching contour layer of the matching image and the matching contour layer of the target image to obtain the contour matching result;

[0042] Step 730: Output the contour matching result from the image matching system.

[0043] Specifically, when the object of contour matching is an image, further processing is required to obtain the contour layer of the object image. This is done by connecting to an image matching system to obtain a matching image, which is used to match the target image. The image matching system matches the contours of the target image to obtain the contour matching result. Specifically, the matching image is analyzed, and its matching contour layer is output. The method for obtaining the matching contour layer of the matching image is consistent with the method for obtaining the matching contour layer of the target image. Finally, contour matching is performed between the matching contour layer of the matching image and the matching contour layer of the target image to obtain the contour matching result. This result is then output through the image matching system, completing the contour matching of the target image.

[0044] The method step 700 provided in this application embodiment further includes:

[0045] Step 740: Connect to the image matching system and obtain the contour layer calling instruction;

[0046] Step 750: According to the contour layer calling instruction, a preset matching contour layer is called from the contour database embedded in the image matching system, wherein the preset matching contour is a preset stored edge contour used for matching with the target image;

[0047] Step 760: Perform contour matching between the preset matching contour layer and the matching contour layer of the target image to obtain the contour matching result.

[0048] Specifically, when the object of contour matching is a contour layer, it can be directly compared with the matching contour layer of the target image. By accessing the image matching system, a contour layer invocation command is obtained. Based on this command, a preset matching contour layer is retrieved from the contour database embedded in the image matching system. This preset matching contour layer consists of pre-stored edge contours from historical contour matching attempts with the target image. Subsequently, this preset matching contour layer is retrieved, and contour matching is performed between it and the matching contour layer of the current target image to obtain the contour matching result, thus completing the contour matching of the target image.

[0049] like Figure 3 As shown, the method step 500 provided in this embodiment further includes:

[0050] Step 510: Determine the grayscale gradient region by analyzing the edge contour of the grayscale converted image;

[0051] Step 520: Obtain a grayscale gradient layer by dividing the grayscale gradient region into grayscale gradients;

[0052] Step 530: Determine the grayscale judgment threshold based on the grayscale gradient layer;

[0053] Step 540: Output the grayscale range based on the grayscale determination threshold;

[0054] Step 550: Perform grayscale analysis using the grayscale layer range as the contour edge determination area to determine the pixel coordinates of the contour layer.

[0055] Specifically, grayscale gradient regions are determined by analyzing the edge contours of the grayscale converted image. Since the image edge contours in a grayscale image exhibit significant changes in grayscale values, the grayscale gradient regions of the image contour edges are identified through edge contour analysis; these regions are then distributed within the grayscale gradient regions. Subsequently, grayscale gradient regions are divided into grayscale gradient layers based on the grayscale value gradient, resulting in grayscale gradient layers that represent the characteristics of grayscale value changes. A grayscale judgment threshold is then determined based on these grayscale gradient layers. This threshold is obtained from the grayscale value data in the grayscale gradient layers. For example, if the grayscale value ranges from 0 to 255, the grayscale judgment threshold is the median of these values. The grayscale judgment threshold can be determined based on the edge grayscale change characteristics of the image, and it is used to distinguish image edge regions. Finally, based on the grayscale judgment threshold, a grayscale layer interval is output, which represents the grayscale region of the target contour edge obtained through the grayscale judgment threshold. Finally, grayscale analysis is performed on the grayscale layer as the contour edge determination area to determine the contour layer pixel coordinates, where the contour layer pixel coordinates are the contour edge pixel coordinates of the target image.

[0056] The method step 550 provided in this embodiment further includes:

[0057] Step 551: By analyzing the gray values ​​in the gray level range, multiple gray level sub-ranges are obtained, wherein the multiple gray level sub-ranges are subsets of the gray level range;

[0058] Step 552: Perform contour sharpness analysis based on the multiple grayscale sub-intervals to obtain the optimal grayscale sub-interval with the highest contour sharpness among the multiple grayscale sub-intervals;

[0059] Step 553: Determine the pixel coordinates of the contour layer based on the optimal grayscale sub-interval.

[0060] Specifically, by analyzing the grayscale values ​​within the grayscale layer range, multiple grayscale sub-ranges are obtained. When obtaining these sub-ranges, the grayscale values ​​within the grayscale layer range are evenly divided, and then the multiple sub-ranges are obtained based on the even division results. These multiple sub-ranges are subsets of the grayscale layer range. Subsequently, contour sharpness analysis is performed on these sub-ranges to obtain the optimal sub-range with the highest contour sharpness. Since a sharper edge display corresponds to a greater variation in pixel values ​​and a larger standard deviation of pixel value ranges, the standard deviation of pixel values ​​in each range can be obtained during contour sharpness analysis. The range with the largest standard deviation is the optimal sub-range with the highest contour sharpness. Alternatively, the optimal sub-range with the highest contour sharpness can be obtained directly through human observation. Based on the optimal sub-range, the pixel coordinates of the contour layer are determined, thus obtaining the pixel coordinates of the contour edge of the target image.

[0061] In summary, the method provided in this application acquires a target image by connecting to an image acquisition device terminal. Image edge extraction is performed on the target image to obtain image edge features. These features are then input into a cloud processor, where they are analyzed based on a contour extraction model linked to the cloud processor to obtain the contour extraction result. The contour extraction result is then input into a grayscale conversion module for grayscale conversion, outputting a grayscale converted image. Grayscale layer analysis is performed on the grayscale converted image to determine the pixel coordinates of the contour layer. Using these contour layer pixel coordinates, a matching contour layer of the target image is output, achieving contour matching. By processing the contour extraction result obtained through image edge extraction to obtain the contour layer pixel coordinates, accurate acquisition of the image contour is achieved. Furthermore, by setting an image quality detection model, the acquired image undergoes quality detection, ensuring that the target image meets predetermined requirements, further improving the image contour extraction accuracy and the target image contour matching accuracy. This solves the technical problem in existing technologies where low image contour extraction accuracy leads to low accuracy in object recognition based on contours.

[0062] Example 2

[0063] Based on the same inventive concept as the method for improving machine vision image contour matching in the foregoing embodiments, such as Figure 4 As shown, this application provides a system for improving the contour matching accuracy of machine vision images. The system is communicatively connected to a cloud processor, and the system includes:

[0064] Image acquisition module 11 is used to connect to the image acquisition device terminal to acquire the target image to be processed, wherein the target image is an image that needs to be contour extracted;

[0065] Image edge feature acquisition module 12 is used to extract image edges from the target image and acquire image edge features;

[0066] The contour extraction module 13 is used to input the image edge features into the cloud processor, analyze them according to the contour extraction model linked into the cloud processor, and obtain the contour extraction result output by the contour extraction model.

[0067] Grayscale conversion module 14 is used to input the contour extraction result into the grayscale conversion module, perform grayscale conversion on the contour extraction result based on the grayscale conversion module, and output a grayscale converted image;

[0068] The contour layer pixel coordinate acquisition module 15 is used to determine the contour layer pixel coordinates by performing grayscale layer analysis on the grayscale converted image;

[0069] The contour matching module 16 is used to output the matching contour layer of the target image using the pixel coordinates of the contour layer, so as to realize contour matching.

[0070] Furthermore, the contour matching module 16 is also used for:

[0071] Access an image matching system to obtain a matching image, wherein the matching image is an image used to match the target image;

[0072] The matching image is analyzed to output the matching contour layer of the matching image;

[0073] The matching contour layer of the matching image is matched with the matching contour layer of the target image to obtain the contour matching result.

[0074] The contour matching result is output by the image matching system.

[0075] Furthermore, the contour matching module 16 is also used for:

[0076] Connect to the image matching system to obtain contour layer calling instructions;

[0077] According to the contour layer calling instruction, a preset matching contour layer is called from the contour database embedded in the image matching system, wherein the preset matching contour is a preset stored edge contour used for matching with the target image;

[0078] The preset matching contour layer is matched with the matching contour layer of the target image to obtain the contour matching result.

[0079] Furthermore, the contour layer pixel coordinate acquisition module 15 is also used for:

[0080] By analyzing the edge contours of the grayscale converted image, the grayscale gradient area is determined;

[0081] A gray-level gradient layer is obtained by dividing the gray-level gradient region into gray-level gradients.

[0082] Based on the grayscale gradient layer, determine the grayscale judgment threshold;

[0083] Based on the grayscale determination threshold, output the grayscale layer range;

[0084] Gray-level analysis is performed using the gray-level range as the contour edge determination area to determine the pixel coordinates of the contour layer.

[0085] Furthermore, the contour layer pixel coordinate acquisition module 15 is also used for:

[0086] By analyzing the gray values ​​within the gray level range, multiple gray sub-ranges are obtained, wherein the multiple gray sub-ranges are subsets of the gray level range;

[0087] Based on the multiple grayscale sub-intervals, an outline sharpness analysis is performed to obtain the optimal grayscale sub-interval with the highest outline sharpness among the multiple grayscale sub-intervals.

[0088] The pixel coordinates of the contour layer are determined based on the optimal grayscale sub-interval.

[0089] Furthermore, the image acquisition module 11 is also used for:

[0090] The target image is input into the image quality detection model, and the quality detection result is obtained according to the image quality detection model, wherein the quality detection result is used to identify the quality level of the target image;

[0091] If the quality inspection result fails, a second image acquisition instruction is obtained;

[0092] According to the image secondary acquisition instruction, the target image is replaced and covered until the quality detection result is that the detection is passed.

[0093] Furthermore, the image acquisition module 11 is also used for:

[0094] The target image is analyzed according to the image quality detection model to obtain image sharpness and color level correlation.

[0095] Based on the image sharpness and the color level correlation as dual-value conditions, a dual-value logic decision-maker is used for analysis, and a decision result is output, wherein the decision result includes pass and fail;

[0096] Based on the decision result, the quality inspection result is obtained.

[0097] The above-described Embodiment 2 is used to execute the method as described in Embodiment 1. Its execution principle and basis can be obtained from the content described in Embodiment 1, and will not be elaborated further here. Although this application has been described in conjunction with specific features and embodiments, this application is not limited to the exemplary embodiments described herein. Based on the embodiments of this application, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application, and the content obtained in this way also falls within the protection scope of this application.

Claims

1. A method for improving contour matching accuracy in machine vision images, characterized in that, The method is applied to an image recognition intelligent processing system, which is communicatively connected to a cloud processor, and the method includes: Connect to the image acquisition device terminal to acquire the target image to be processed, wherein the target image is an image that needs to be contour extracted; Image edge extraction is performed on the target image to obtain image edge features; The image edge features are input into the cloud processor, and the contour extraction results are obtained by analyzing the contour extraction model linked into the cloud processor. The contour extraction result is input into the grayscale conversion module, and the grayscale conversion is performed on the contour extraction result based on the grayscale conversion module to output the grayscale converted image; By performing grayscale layer analysis on the grayscale converted image, the pixel coordinates of the contour layer are determined; Using the pixel coordinates of the contour layer, the matching contour layer of the target image is output for contour matching. The system is connected to an image matching system, and the method further includes: Access an image matching system to obtain a matching image, wherein the matching image is an image used to match the target image; The matching image is analyzed to output the matching contour layer of the matching image; The matching contour layer of the matching image is matched with the matching contour layer of the target image to obtain the contour matching result. The contour matching result is output by the image matching system; The method further includes performing grayscale layer analysis on the grayscale converted image: By analyzing the edge contours of the grayscale converted image, the grayscale gradient area is determined; A gray-level gradient layer is obtained by dividing the gray-level gradient region into gray-level gradients. Based on the grayscale gradient layer, determine the grayscale judgment threshold; Based on the grayscale determination threshold, output the grayscale layer range; Gray-level analysis is performed using the gray-level range as the contour edge determination area to determine the pixel coordinates of the contour layer; The method further includes performing grayscale analysis using the grayscale layer range as the contour edge determination region: By analyzing the gray values ​​within the gray level range, multiple gray sub-ranges are obtained, wherein the multiple gray sub-ranges are subsets of the gray level range; Based on the multiple grayscale sub-intervals, an outline sharpness analysis is performed to obtain the optimal grayscale sub-interval with the highest outline sharpness among the multiple grayscale sub-intervals. The pixel coordinates of the contour layer are determined based on the optimal grayscale sub-interval.

2. The method as described in claim 1, characterized in that, The method for obtaining the contour matching result further includes: Connect to the image matching system to obtain contour layer calling instructions; According to the contour layer calling instruction, a preset matching contour layer is called from the contour database embedded in the image matching system, wherein the preset matching contour is a preset stored edge contour used for matching with the target image; The preset matching contour layer is matched with the matching contour layer of the target image to obtain the contour matching result.

3. The method as described in claim 1, characterized in that, After acquiring the target image to be processed, the method further includes: The target image is input into the image quality detection model, and the quality detection result is obtained according to the image quality detection model, wherein the quality detection result is used to identify the quality level of the target image; If the quality inspection result fails, a second image acquisition instruction is obtained; According to the image secondary acquisition instruction, the target image is replaced and covered until the quality detection result is that the detection is passed.

4. The method as described in claim 3, characterized in that, The method further includes: The target image is analyzed according to the image quality detection model to obtain image sharpness and color level correlation. Based on the image sharpness and the color level correlation as dual-value conditions, a dual-value logic decision-maker is used for analysis, and a decision result is output, wherein the decision result includes pass and fail; Based on the decision result, the quality inspection result is obtained.

5. A system for improving contour matching accuracy in machine vision images, characterized in that, The system is used to perform the method according to any one of claims 1-4, the system is communicatively connected to a cloud processor, and the system comprises: An image acquisition module is used to connect to an image acquisition device terminal to acquire a target image to be processed, wherein the target image is an image for which contour extraction is required; The image edge feature acquisition module is used to extract image edges from the target image and acquire image edge features. The contour extraction module is used to input the image edge features into the cloud processor, analyze them according to the contour extraction model linked into the cloud processor, and obtain the contour extraction results output by the contour extraction model. A grayscale conversion module is used to input the contour extraction result into the grayscale conversion module, perform grayscale conversion on the contour extraction result based on the grayscale conversion module, and output a grayscale converted image; The contour layer pixel coordinate acquisition module is used to determine the contour layer pixel coordinates by performing grayscale layer analysis on the grayscale converted image; The contour matching module is used to output the matching contour layer of the target image using the pixel coordinates of the contour layer, thereby achieving contour matching.

Images