Data collection method and system of station unit

Abstract

The invention relates to a data collection method and system for a station unit, and the method can extract the picture information in a first image through receiving the first image, and the picture information comprises one or more of text information, digital information and product appearance information. The picture information is compared with a production inspection target, so that the preliminary process of production inspection of the station unit can be completed on the premise of saving production inspection time and manpower resources. Because the current image modification software is relatively common, in order to improve the authenticity of the first image and improve the production inspection process of the station unit, the second image is received, and the photographing authenticity of the first image can be distinguished by using the second image. Information authenticity of the first image may be verified. Meanwhile, the second image can have the content information of the photographer of the first image, so that the authenticity of the second image can be determined based on a three-dimensional image restoration algorithm.

Description

Technical Field

[0001] The present application relates to the field of image algorithm technology, and in particular to a data collection method and system for a workstation unit. Background Art

[0002] Efficient and high-quality production of products is an important way for factories to pursue economic benefits. In order to achieve efficient production of products, factories often use assembly line work mode, and each workstation unit of the assembly line is independent of each other. In order to achieve high-quality production of products, factories often use inspection to conduct production inspection on each workstation unit.

[0003] The traditional production inspection method of workstation units is time-consuming and human resource-consuming. How to complete the production inspection process of workstation units while saving time and human resources is a current industry difficulty. To this end, it is necessary to propose a data collection method and system for workstation units in view of the defect that the production inspection of traditional workstation units is relatively resource-consuming. Summary of the invention

[0004] Based on this, it is necessary to propose a data collection method and system for a workstation unit to address the defect that production inspection of traditional workstation units consumes more resources.

[0005] The present application provides a data collection method for a workstation unit, comprising:

[0006] receiving a first image and a second image; wherein the content of the first image is production content, and the content of the second image is the content of the first image and an action of a shooter of the first image when shooting the first image;

[0007] Based on the three-dimensional image restoration algorithm, the perspective feature extraction of the second image is realized;

[0008] Invoking an image recognition algorithm to extract the content of the first image from the second image;

[0009] determining whether the content of the first image in the second image matches the target of the production inspection;

[0010] If the content of the first image in the second image does not match the target of production inspection, a prompt message indicating that the inspection process is abnormal is fed back;

[0011] If the content of the first image in the second image matches the target of the production inspection, then further determining whether the perspective feature of the second image matches the three-dimensional image feature in the three-dimensional image restoration algorithm;

[0012] If the perspective features of the second image do not match the three-dimensional image features in the three-dimensional image restoration algorithm, a prompt message indicating that an abnormality has occurred in the inspection process is fed back;

[0013] If the perspective features of the second image match the three-dimensional image features in the three-dimensional image restoration algorithm, a prompt message indicating that the inspection process has been completed is fed back.

[0014] The present application provides a data collection system for a workstation unit, comprising:

[0015] A processor, used to execute the data collection method of the workstation unit;

[0016] A first image collector, communicatively connected to the processor;

[0017] The second image collector is communicatively connected with the processor.

[0018] The present application relates to a data collection method and system for a workstation unit. By receiving a first image, the picture information in the first image can be extracted, and the picture information includes one or more of text information, digital information and product appearance information. Comparing the picture information with the target of production inspection can save time and human resources for production inspection, and complete the preliminary process of production inspection of the workstation unit. Since the current image modification software is relatively common, in order to improve the authenticity of the first image and improve the process of production inspection of the workstation unit, by receiving a second image, the authenticity of the shooting of the first image can be identified by using the second image. Specifically, the second image has the work order image in the first image, the photographer of the first image and other contents. The authenticity of the information of the first image can be verified by the work order image in the second image. At the same time, the second image can have the content information of the photographer of the first image, so based on the three-dimensional image restoration algorithm, the authenticity of the second image itself can be determined. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic flow chart of a data collection method for a workstation unit provided in one embodiment of the present application.

[0020] Figure 2 A structural connection diagram of a data collection system for a workstation unit provided in one embodiment of the present application.

[0021] Reference numerals:

[0022] 100 - processor; 200 - first image collector; 300 - second image collector. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] The present application provides a data collection method for a workstation unit.

[0025] like Figure 1 As shown, in one embodiment of the present application, a data collection method for a workstation unit includes:

[0026] S100, receiving a first image and a second image, wherein the content of the first image is production content, and the content of the second image is the content of the first image and an action of a shooter of the first image when shooting the first image.

[0027] Specifically, the content of the first image is mainly the project work order to be inspected, the product, etc. Production inspection can be implemented using the first image project work order content and product appearance image information.

[0028] For example, a project work order is photographed to obtain an image of the project work order, and a product is photographed to obtain a three-dimensional image. Both of these images can be used as the first image.

[0029] The content of the second image is mainly the content of the first image and the action of the photographer of the first image when shooting the first image.

[0030] For example, photographer A holds camera A1 to shoot a product and obtains the first image. When photographer A holds camera A1 to shoot the product, photographer B holds camera B1 and finds a suitable angle to shoot the action that photographer A is shooting and obtains the second image. Therefore, the second image requires both the content of the first image (i.e. what photographer A shoots), and the shooting action and photographer of the first image (i.e. the action of the photographer of the first image shooting the first image).

[0031] It should be noted that the content of the first image contained in the second image does not need to have too high a definition, as long as it can pass subsequent recognition.

[0032] S200, based on a three-dimensional image restoration algorithm, extracting perspective features of the second image.

[0033] Specifically, since the content of the first image may be a work order of a project to be inspected, in this working state, it is difficult to determine whether the first image is an image captured in real time by using a three-dimensional image restoration algorithm.

[0034] To this end, it is necessary to review the correctness of the content of the first image using the second image containing the content of the first image and the first image taken by the photographer of the first image, and determine the authenticity of the second image based on the real-time three-dimensional spatial state of the photographer of the first image.

[0035] Since the second image covers more content, image recognition is more difficult and the accuracy of the image recognition result is lower by only using the project work order content and product appearance image information of the first image in the second image. By using the second image to verify the project work order content and product appearance image information of the first image, the production inspection process of the entire workstation unit has a higher fault tolerance rate.

[0036] S300, calling an image recognition algorithm to extract content of the first image in the second image.

[0037] Specifically, image recognition algorithms may be used to extract the content of the first image and to extract the content of the first image in the second image.

[0038] Since the content of the first image may contain text content (i.e., words), when text content appears, an image recognition algorithm represented by a convolutional neural network (CNN) can be used to automatically learn and extract the words appearing in the content of the first image by building a deep network structure.

[0039] When the appearance image information of the product appears, the K-means clustering algorithm can be used for image segmentation, which can be used without pre-labeled training data to obtain the appearance image information of the product.

[0040] S400, determining whether the content of the first image in the second image is consistent with the target of production inspection.

[0041] S500: If the content of the first image in the second image does not match the target of the production inspection, a prompt message indicating that an abnormality has occurred in the inspection process is fed back.

[0042] S600: If the content of the first image in the second image matches the target of the production inspection, further determine whether the perspective feature of the second image matches the three-dimensional image feature in the three-dimensional image restoration algorithm.

[0043] S700: If the perspective feature of the second image does not match the three-dimensional image feature in the three-dimensional image restoration algorithm, a prompt message indicating that an abnormality has occurred in the inspection process is fed back.

[0044] S800: If the perspective feature of the second image matches the three-dimensional image feature in the three-dimensional image restoration algorithm, a prompt message indicating that the inspection process has been completed is fed back.

[0045] This embodiment relates to a data collection method for a workstation unit. By receiving a first image, the picture information in the first image can be extracted. The picture information includes one or more of text information, digital information and product appearance information. Comparing the picture information with the target of production inspection can save time and human resources for production inspection, and complete the preliminary process of production inspection of the workstation unit. Since the image modification software is relatively common at present, in order to improve the authenticity of the first image and improve the process of production inspection of the workstation unit, by receiving a second image, the authenticity of the shooting of the first image can be identified by using the second image. Specifically, the second image has the work order image in the first image, the photographer of the first image and other contents. The authenticity of the information of the first image can be verified by the work order image in the second image. At the same time, the second image can have the content information of the photographer of the first image, so based on the three-dimensional image restoration algorithm, the authenticity of the second image itself can be determined.

[0046] In one embodiment of the present application, S100 includes:

[0047] S111, receiving a first image and a second image.

[0048] Specifically, the first image is shot simultaneously with the second image to ensure that the content of the second image has the characteristics of a three-dimensional space.

[0049] S112, performing illumination correction on the first image and the second image respectively to obtain a corrected first image and a corrected second image.

[0050] Specifically, image preprocessing is a series of processing steps aimed at improving the quality of image data and providing more accurate input data for subsequent image analysis and pattern recognition processes. These processing methods include illumination correction, blur processing and denoising, image pyramids, edge extraction and enhancement, etc.

[0051] S113, performing Gaussian filtering on the corrected first image and the corrected second image respectively to obtain a first initial image and a second initial image.

[0052] Specifically, the weight of the Gaussian function is used to perform weighted averaging on the neighborhood pixels, effectively removing the normally distributed noise, which is widely used in the noise reduction process in image processing.

[0053] S114, binarizing the first initial image to obtain a grayscale image of the first image.

[0054] Specifically, the image is converted into a binary image with only two grayscale values ​​for image segmentation and object extraction. The binary grayscale value can be used for edge extraction and enhancement, and can be used to divide the grayscale image of the two-dimensional plane into blocks in the two-dimensional plane.

[0055] S115, binarizing the second initial image to obtain a grayscale image of the second image.

[0056] This embodiment involves image preprocessing, which is an important part of computer vision tasks. Image quality is improved by means of illumination correction, blur processing and denoising, image pyramid, edge extraction and enhancement, etc., making the image more suitable for further analysis and processing. In this embodiment, an appropriate preprocessing method can be selected according to specific tasks, such as text recognition or product geometric appearance recognition, to ensure the accuracy of the final result.

[0057] More specifically, when text appears, an image recognition algorithm represented by a convolutional neural network (CNN) can be used to automatically learn and extract the text that appears in the content of the first image by building a deep network structure. In this case, the preprocessing method of blurring and denoising can improve the accuracy of text recognition.

[0058] When the appearance image information of the product appears, the K-means clustering algorithm can be used for image segmentation without pre-labeled training data to obtain the appearance image information of the product. In this case, the edge extraction and enhancement preprocessing method can improve the accuracy of the product appearance image information.

[0059] In one embodiment of the present application, S100 further includes:

[0060] S121, extracting edges of the grayscale image, and selecting a plurality of core information blocks from the grayscale image.

[0061] Specifically, the image is converted into a binary image with only two grayscale values, namely white and gray, for image segmentation and object extraction. Binary grayscale values ​​can be used for edge extraction and enhancement, and can be used for grayscale images in a two-dimensional plane to extract block edges in a two-dimensional plane. Under this edge extraction method, the area with grayscale values ​​has distinguishable edges from the pure white area, and the core information block is the gray extraction object. In fact, the grayscale image can also be segmented in layers according to the grayscale value, and each grayscale layer is vector-connected. Layered segmentation adds a dimension in two-dimensional space, and this dimension is the grayscale value. The layered segmented grayscale image can be used for edge extraction and enhancement in a three-dimensional data system.

[0062] S122, select a core information block.

[0063] Specifically, the segmented grayscale image is a grayscale image c of a two-dimensional plane.

[0064] S123, based on the VGG deep learning neural network, extracts information from each core information block.

[0065] Specifically, the information in the core information block includes one or more of text information, digital information and product appearance information.

[0066] S123a, establishing a filter for the grayscale image, and using the grayscale image as the original image to be convolved.

[0067] Specifically, a color / grayscale image has multiple RGB channels, and the image can be convolved using multiple filters with the same number of channels.

[0068] S123b, generating a secondary convolution image based on the moving step size of the filter.

[0069] Specifically, the distance that the filter moves parallely in the image to be convolved is the moving step length of the filter.

[0070] The filter can move parallel to the image to be convolved in two directions. The segmented grayscale image is a grayscale image in a two-dimensional plane.

[0071] S123c, using a pooler to pool the secondary convolution image.

[0072] Specifically, the pooler can have two methods: maximum pooling and average pooling. The pooler does not participate in the convolution calculation, and the pooler is mainly used for dimensionality reduction of the image to be convolved.

[0073] S123d, returning to the filter for establishing the grayscale image until the dimension of the secondary convolution image reaches a dimension threshold.

[0074] S123e, fully connect the secondary convolution image that reaches the dimension threshold.

[0075] Specifically, the full connection process can actually be the matrixization of the convolution result of the previous level image to be convolved, and based on the matrixization result, the next level image to be convolved can be generated.

[0076] S123f, obtain the information of the core information block in the grayscale image.

[0077] Specifically, by using the results of multiple matrixing, the dimension of the original image to be convolved can be reduced to a level that can be output.

[0078] More specifically, by outputting a result with high credibility, the information in the grayscale image can be obtained.

[0079] S124, returning to the step of selecting a core information block until all core information blocks are selected.

[0080] In one embodiment of the present application, S200 includes:

[0081] S211, based on the grayscale image of the second image, outlining the grayscale image of the second image to obtain a plurality of contour images.

[0082] Specifically, the Sobel operator detects contours by calculating the gradient of the grayscale values ​​around each pixel in the image. It uses two 3x3 kernels to convolve the original image in the horizontal and vertical directions respectively, thereby obtaining two components of the gradient, and finally obtains the edge image by merging these two components.

[0083] S212, select a contour map.

[0084] S213, forming a perspective line of the outline image.

[0085] Specific:

[0086] S213a, calling a second image.

[0087] S213b, forming a multi-layer grayscale image based on the division of grayscale values.

[0088] S213c, convolve the grayscale image of each layer in the horizontal and vertical directions.

[0089] S213d, obtain two components of the gradient in the horizontal direction and the vertical direction.

[0090] S213e, based on the grayscale image of each layer, merge the two components of the gradient in the horizontal direction and the vertical direction.

[0091] S213f, obtain edge image.

[0092] S213g, define the edge image as a set of contour images of the second image.

[0093] Specifically, the grayscale image can also be segmented in layers according to the grayscale value, and each grayscale layer can be vector-connected. The layered segmentation adds a dimension in the two-dimensional space, which is the grayscale value. The layered segmented grayscale image can be used for edge extraction and enhancement in the three-dimensional data system.

[0094] S213z, select a contour map from the set of contour maps of the second image.

[0095] S213y, extract multiple straight line segments of the contour image.

[0096] S213x, extend each straight line segment to form a perspective line of the contour image.

[0097] S213w, returning a contour image in the set of contour images selected for the second image, until all contour images are selected.

[0098] Specifically, the Sobel operator emphasizes edges by calculating the gradient of each pixel in the image, especially those areas where the grayscale changes significantly. By setting an appropriate threshold, straight line segments can be extracted from the Sobel-processed image.

[0099] It is worth mentioning that in two-dimensional space, straight lines parallel to the base coordinates of the two-dimensional space will be eliminated, and the remaining straight line segments can be used to determine the focus of the perspective line.

[0100] In the current image collector, the focus of the perspective line may not be unique, and the distance between the focus of the perspective line is specific. The distance between the focus of the perspective line is one of the three-dimensional image features in the three-dimensional image restoration algorithm.

[0101] S214, returning to the step of selecting a contour map until all contour maps are selected.

[0102] S221, select a perspective line of a contour drawing.

[0103] S222: Determine the focus of the perspective line based on the perspective line of the outline image.

[0104] S223, defining the focus of the perspective line as a feature of the second image.

[0105] S224, returning to the step of selecting a perspective line of a contour drawing until all perspective lines of the contour drawings are selected.

[0106] Specifically, the focus of the perspective line of the images captured by the image collector in real time can reflect the actual three-dimensional spatial position of the object in the image, while it is difficult for the focus of the perspective line of the modified image to reflect the actual three-dimensional spatial position of the object in the image.

[0107] Simply, the focal point of the perspective line of the real-time captured image cannot coincide with the focal point of the perspective line of the modified image.

[0108] In one embodiment of the present application, S600 includes:

[0109] S611: Select a grayscale image of a second image.

[0110] S612, determining the geometric center of the grayscale image of the second image in the two-dimensional plane, and taking the geometric center of the grayscale image of the second image in the two-dimensional plane as a three-dimensional image feature.

[0111] S613, select a focal point of a perspective line.

[0112] S614: Determine whether the focus of the perspective line coincides with the geometric center of the second image.

[0113] S615, if the focus of the perspective line does not match the geometric center of the second image, it is determined that the features of the second image do not match the three-dimensional image features in the three-dimensional image restoration algorithm, and the process of selecting a grayscale image of the second image is returned until all grayscale images of the second image are selected.

[0114] S616: If the focus of the perspective line coincides with the geometric center of the second image, the process returns to selecting a focus of a perspective line until all the focuses of the perspective lines are selected.

[0115] S617, when all the focal points of the perspective lines are selected, it is determined whether all the focal points of the selected grayscale image of the second image coincide with the geometric center of the second image.

[0116] S618, if the focal points of all perspective lines of the grayscale image of the selected second image coincide with the geometric center of the second image, it is determined that the features of the second image coincide with the three-dimensional image features in the three-dimensional image restoration algorithm, and the process of selecting a grayscale image of a second image is returned until all grayscale images of the second image are selected.

[0117] S619, if there is at least one perspective line whose focus does not match the geometric center of the second image, it is determined that the features of the second image do not match the three-dimensional image features in the three-dimensional image restoration algorithm, and the process of selecting a grayscale image of the second image is returned until all grayscale images of the second image are selected.

[0118] The present application provides a data collection system for a workstation unit.

[0119] like Figure 2 As shown, in one embodiment of the present application, a data collection system for a workstation unit includes a processor 100 , a first image collector 200 , and a second image collector 300 .

[0120] The processor 100 is used to execute the data collection method of the workstation unit.

[0121] The first image collector 200 is in communication connection with the processor 100 .

[0122] The second image collector 300 is in communication connection with the processor 100 .

[0123] This embodiment relates to a data collection system for a workstation unit. The processor 100 can extract the picture information in the first image by receiving the first image captured by the first image collector 200. The picture information includes one or more of text information, digital information and product appearance information. The comparison of the picture information with the target of production inspection can save the time and human resources of production inspection, and complete the preliminary process of production inspection of the workstation unit. Since the current image modification software is relatively common, in order to improve the authenticity of the first image and improve the process of production inspection of the workstation unit, by receiving the second image captured by the second image collector 300, the authenticity of the shooting of the first image can be identified by the second image. Specifically, the second image has the work order image in the first image, the photographer of the first image and other contents. The authenticity of the information of the first image can be verified by the work order image in the second image. At the same time, the second image can have the content information of the photographer of the first image, so based on the three-dimensional image restoration algorithm, the authenticity of the second image itself can be determined.

[0124] The technical features of the above-described embodiments may be arbitrarily combined, and the execution order of the method steps is not limited. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data collection method for a workstation unit, characterized in that: include: receiving a first image and a second image; wherein the content of the first image is production content, and the content of the second image is the content of the first image and an action of a shooter of the first image when shooting the first image; Based on the three-dimensional image restoration algorithm, the perspective feature extraction of the second image is realized; Invoking an image recognition algorithm to extract the content of the first image from the second image; determining whether the content of the first image in the second image matches the target of the production inspection; If the content of the first image in the second image does not match the target of production inspection, a prompt message indicating that the inspection process is abnormal is fed back; If the content of the first image in the second image matches the target of the production inspection, determining whether the perspective feature of the second image matches the three-dimensional image feature in the three-dimensional image restoration algorithm; If the perspective features of the second image do not match the three-dimensional image features in the three-dimensional image restoration algorithm, a prompt message indicating that an abnormality has occurred in the inspection process is fed back; If the perspective features of the second image match the three-dimensional image features in the three-dimensional image restoration algorithm, a prompt message indicating that the inspection process has been completed is fed back.

2. The data collection method of the workstation unit according to claim 1, characterized in that: The receiving of the first image and the second image comprises: receiving a first image and a second image; Performing illumination correction on the first image and the second image respectively to obtain a corrected first image and a corrected second image; Performing Gaussian filtering on the corrected first image and the corrected second image respectively to obtain a first initial image and a second initial image; Binarizing the first initial image to obtain a grayscale image of the first image; The second initial image is binarized to obtain a grayscale image of the second image.

3. The data collection method of the workstation unit according to claim 2, characterized in that: The receiving of the first image and the second image further includes: Extract the edge of the grayscale image and select multiple core information blocks from the grayscale image; Choose a core information block; Based on the VGG deep learning neural network, extract the information in each core information block; the information in the core information block includes one or more of text information, digital information and product appearance information; Return to selecting a core information block until all core information blocks are selected.

4. The data collection method of the workstation unit according to claim 3, characterized in that: The VGG deep learning neural network is used to extract information from each core information block, including: Establishing a filter of the grayscale image, and taking the grayscale image as the original convolution image; Generate a secondary convolution image based on the filter's moving step size; Use the pooler to pool the secondary convolution image; Returning the filter for creating the grayscale image until the dimension of the secondary convolution image reaches a dimension threshold; Fully connect the secondary convolution images that reach the dimension threshold; Get the information of the core information block in the grayscale image.

5. The data collection method of the workstation unit according to claim 4, characterized in that: The method of extracting the perspective feature of the second image based on the three-dimensional image restoration algorithm includes: Based on the grayscale image of the second image, outlining the grayscale image of the second image to obtain a plurality of contour images; Select a contour map; the perspective lines that form this outline; Return to Select a Contour until all contours are selected.

6. The data collection method of the workstation unit according to claim 5, characterized in that: The method of extracting the perspective feature of the second image based on the three-dimensional image restoration algorithm further includes: Choose a perspective line for a contour drawing; Based on the perspective line of the outline image, determining the focal point of the perspective line; defining a focal point of the perspective line as a perspective feature of the second image; Return to the step of selecting a perspective line of a contour drawing until all perspective lines of the contour drawings are selected.

7. The data collection method of the workstation unit according to claim 6, characterized in that: The perspective lines forming the outline diagram include: Calling a second image; Based on the division of grayscale values, a multi-layer grayscale image is formed; Convolve the grayscale image of each layer in the horizontal and vertical directions; Get the two components of the gradient in the horizontal and vertical directions; Based on the grayscale image of each layer, merge the two components of the gradient in the horizontal and vertical directions; Obtaining an edge image; Define the edge image as the collection of contour maps of the second image.

8. The data collection method of the workstation unit according to claim 7, characterized in that: The perspective lines forming the outline map also include: selecting a contour map from the set of contour maps of the second image; Extract multiple straight line segments of the contour image; Extend each straight line segment to form the perspective line of the outline; Return a contour map in the set of contour maps of the selected second image until all contour maps are selected.

9. The data collection method of the workstation unit according to claim 8, characterized in that: The determining whether the feature of the second image is consistent with the feature of the three-dimensional image in the three-dimensional image restoration algorithm includes: Selecting a grayscale image of a second image; Determine the geometric center of the grayscale image of the second image in the two-dimensional plane, and use the geometric center of the grayscale image of the second image in the two-dimensional plane as a three-dimensional image feature; Choose a focal point for a perspective line; determining whether the focus of the perspective line coincides with the geometric center of the second image; If the focus of the perspective line does not coincide with the geometric center of the second image, it is determined that the features of the second image do not coincide with the three-dimensional image features in the three-dimensional image restoration algorithm, and the process returns to selecting a grayscale image of a second image until all grayscale images of the second image are selected.

10. The data collection method of the workstation unit according to claim 9, characterized in that: The step of determining whether the feature of the second image is consistent with the feature of the three-dimensional image in the three-dimensional image restoration algorithm further includes: If the focus of the perspective line coincides with the geometric center of the second image, returning to the step of selecting a focus of a perspective line until the focus of all perspective lines is selected; When all the focal points of the perspective lines are selected, it is determined whether all the focal points of the selected grayscale image of the second image coincide with the geometric center of the second image; If the focal points of all perspective lines of the grayscale image of the selected second image coincide with the geometric center of the second image, it is determined that the feature of the second image coincides with the three-dimensional image in the three-dimensional image restoration algorithm, and the grayscale image of the selected second image is returned until all the grayscale images of the second image are selected; if there is at least one focal point of the perspective line that does not coincide with the geometric center of the second image, it is determined that the feature of the second image does not coincide with the three-dimensional image in the three-dimensional image restoration algorithm, and the grayscale image of the selected second image is returned until all the grayscale images of the second image are selected.

Images