Industrial defect detection method, system, equipment, medium and program product
By enlarging, segmenting, enhancing processing and iteratively enhancing the original images of industrial products, minor defects are highlighted and image stitching and shrinking are performed, the problem of difficult detection of minor defects in the prior art is solved, and more accurate and comprehensive defect detection is achieved.
Patent Information
- Application Number
- CN202411748906.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-02
- Publication Date
- 2025-03-14
AI Technical Summary
Existing industrial defect detection technology is difficult to effectively detect minor defects, such as scratches and highlights, and it is easy to cause missed inspection.
By amplifying, segmenting, enhancing and edge contour extraction of the original images of industrial products, the fine texture features of the regional contour image are obtained, and contour iteratively enhance the processing to highlight minor defects. Then, the contour enhancement image is stitched and reduced to make its size the same as the original image, and finally defect detection is performed simultaneously in the defect detection system.
Effectively prevent missed detection of minor defects and improve the accuracy and comprehensiveness of defect detection.
Smart Images
Figure CN119624920A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of industrial defect detection, and in particular to an industrial defect detection method, system, equipment, medium and program product. Background Art
[0002] With the rapid development of industrial manufacturing, automated production lines have become the main production method for many factories. However, due to equipment aging, operational errors and other reasons, industrial products often have various forms of defects during the production process. If these defects cannot be detected and handled in time, the quality of panel production will be reduced.
[0003] At present, industrial product defect detection often uses an automatic defect detection system (ADC system) for defect detection, such as industrial panels and other industrial products. In the process of industrial product defect target detection, the original image of the industrial product is first enhanced, and then the original image processed by image enhancement is input into the automatic defect classification system for defect detection. However, for some minor defects such as scratches and bright spots, it is difficult for the automatic defect detection system to detect them, and it is easy to miss the problem. Summary of the invention
[0004] In order to solve the problem that the existing industrial defect detection technology is prone to missing detection of minor defects, the embodiments of the present invention provide an industrial defect detection method, system, device, medium and program product.
[0005] In a first aspect, an embodiment of the present invention provides an industrial defect detection method, the method comprising the following steps:
[0006] Perform image magnification and image segmentation processing on the original image of the industrial product to obtain a region segmentation image;
[0007] Performing image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image;
[0008] Performing iterative contour enhancement processing on the regional contour image to obtain a contour enhanced image;
[0009] Perform image stitching and image reduction processing on the contour enhanced image to obtain the original enhanced image of the industrial product;
[0010] The original image and the original enhanced image of the industrial product are associated and input into the defect detection system for defect detection to obtain defect location results and defect supplement location results;
[0011] The defect location results and defect supplement location results are comprehensively judged to obtain the final defect detection results.
[0012] In the above embodiment, the present invention can obtain more detailed texture features of the regional contour image by performing image magnification, image segmentation, image enhancement processing and edge contour extraction processing on the original image, and then perform contour iterative enhancement processing on the regional contour image, which can enhance the size of the grayscale value corresponding to the slight defect, thereby highlighting the small and slight defects and reducing the size of the grayscale value corresponding to the deep regular background image, thereby weakening the deep and regular background; then the contour enhanced image is subjected to image stitching and image reduction processing, so that the size of the original enhanced image is the same as the original image size, and finally, defect detection is performed on the original image and the original enhanced image at the same time, which can effectively prevent the situation where slight defects are easily missed.
[0013] As some optional implementations of the present application, the process of image enlargement and image segmentation processing on the original image of the industrial product is as follows:
[0014] Presetting a magnification of the original image, performing image magnification processing on the original image based on the magnification factor, and performing image sharpening processing and image denoising processing on the original image after the image magnification processing to obtain the original magnified image;
[0015] A segmentation strategy of the original enlarged image is preset, and image segmentation processing is performed on the original enlarged image based on the segmentation strategy, and boundary optimization processing is performed on the original enlarged image after image segmentation processing to obtain a region segmentation image.
[0016] In the above embodiment, the present invention performs image magnification and image segmentation processing on the original image to obtain more detailed texture features, and because the size of the image remains unchanged after segmentation, the original defect detection related parameters and algorithms can still be used, which is more universal.
[0017] As some optional implementations of the present application, the process of performing image enhancement processing and edge contour extraction processing on the region segmentation image is as follows:
[0018] Performing forward Fourier transform on the region segmentation image to convert the region segmentation image from the time domain to the frequency domain;
[0019] Performing high-frequency filtering and inverse Fourier transform on the regional segmented image in the frequency domain to obtain an initial enhanced image;
[0020] The Laplacian operator is used to perform texture and edge detection on the initial enhanced image to obtain a region contour image.
[0021] In the above embodiment, the present invention can obtain clearer contour information by performing image enhancement processing and edge contour extraction processing on the region segmentation image.
[0022] As some optional implementations of the present application, the process of iteratively enhancing the contour of the regional contour image is as follows:
[0023] Presetting adjustment parameters, and calculating conversion coefficients of the region contour image based on the adjustment parameters;
[0024] The grayscale value of the regional contour image is converted into a triangular space based on the conversion coefficient to obtain an image matrix corresponding to the regional contour image;
[0025] Calculate the maximum and minimum values of the image matrix, and scale the image matrix to the grayscale space based on the maximum and minimum values of the image matrix;
[0026] Preset the grayscale threshold, grayscale increase value, grayscale reduction value and number of iterative enhancements for iterative enhancement;
[0027] Within the range of iterative enhancement times, the grayscale value of the regional contour image in the grayscale space is threshold checked based on the grayscale threshold. If the grayscale value is greater than the grayscale threshold, the grayscale value is subtracted according to the grayscale reduction value, and the remaining grayscale value is added according to the grayscale increase value, so as to iteratively obtain the contour enhanced image of the industrial product.
[0028] In the above embodiment, the present invention can enhance the size of the grayscale value corresponding to slight defects by performing iterative contour enhancement processing on the regional contour image, thereby highlighting the small and slight defects, reducing the size of the grayscale value corresponding to the deep regular background image, and thus weakening the deep and regular background.
[0029] As some optional implementations of the present application, the process of image stitching and image reduction processing on the contour enhanced image is as follows:
[0030] Performing image stitching processing on the contour enhanced image in reverse according to the segmentation strategy, and performing boundary fusion on the contour enhanced image processed by image stitching to obtain the original stitching image;
[0031] The original stitched image is reversely reduced based on the magnification of the original image, and the image is sharpened and denoised to obtain the original enhanced image of the industrial product.
[0032] In the above embodiment, the present invention performs image stitching and image reduction processing on the edge enhanced image so that the size of the original enhanced image is the same as the size of the original image.
[0033] As some optional implementations of the present application, the process of associating the original image of the industrial product and the original enhanced image and inputting them into the defect detection system for defect detection is as follows:
[0034] Correlating an original image of an industrial product and a corresponding original enhanced image;
[0035] The mutually related original image and the original enhanced image are input into a pre-trained defect detection system for detection, and the pre-trained defect detection system outputs a defect location result and a defect supplement location result respectively; wherein the defect location result and the defect supplement location result respectively include a defect position and a defect category.
[0036] In the above-mentioned embodiment, the present invention can obtain severe defect information by inputting the original image into a pre-trained defect detection system; and can obtain slight defect information by inputting the original enhanced image into a pre-trained defect detection system.
[0037] As some optional implementation methods of the present application, the process of comprehensively judging the defect location results and the defect supplement location results is as follows:
[0038] Determine whether there is overlap in defect positions based on the defect location results and defect supplement location results;
[0039] If there is overlap in the defect positions, continue to determine whether the corresponding defect types are the same. If the corresponding defect types are the same, retain the corresponding defect location results; if the defect types are different, retain the corresponding defect location results and defect supplement location results to obtain the final defect detection result;
[0040] If there is no overlap in defect positions, the corresponding defect location results and defect supplement location results are directly retained to obtain the final defect detection result.
[0041] In the above embodiment, the present invention can effectively remove duplication by performing comprehensive verification on serious defect information and minor defect information.
[0042] In a second aspect, the present invention provides an industrial defect detection system, the functions of the system corresponding to the steps of the method, the system comprising:
[0043] An image segmentation unit, wherein the original image segmentation unit is used to perform image magnification processing and image segmentation processing on the original image of the industrial product to obtain a region segmentation image;
[0044] A contour acquisition unit, the contour acquisition unit is used to perform image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image;
[0045] A contour enhancement unit, wherein the contour enhancement unit is used to perform contour iterative enhancement processing on the regional contour image to obtain a contour enhanced image;
[0046] An image stitching unit, the image stitching unit is used to perform image stitching processing and image reduction processing on the contour enhanced image to obtain an original enhanced image of the industrial product;
[0047] An associative positioning unit, the associative positioning unit is used to associate the original image and the original enhanced image of the industrial product and input them into the defect detection system for defect detection to obtain a defect positioning result and a defect supplementary positioning result;
[0048] The comprehensive determination unit is used to make a comprehensive determination on the defect location result and the defect supplement location result to obtain a final defect detection result.
[0049] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the industrial defect detection method.
[0050] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the industrial defect detection method when executed by a processor.
[0051] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which implements the industrial defect detection method when executed by a processor.
[0052] The beneficial effects of the present invention are as follows: the present invention can obtain more detailed texture features of the regional contour image by performing image magnification, image segmentation, image enhancement processing and edge contour extraction processing on the original image, and then perform contour iterative enhancement processing on the regional contour image, which can enhance the size of the grayscale value corresponding to the slight defect, thereby highlighting the small and slight defects, and reduce the size of the grayscale value corresponding to the deep regular background image, thereby weakening the deep and regular background; then the contour enhanced image is subjected to image stitching and image reduction processing, so that the size of the original enhanced image is the same as the size of the original image, and finally, defect detection is performed on the original image and the original enhanced image at the same time, which can effectively prevent the situation where slight defects are easily missed. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0054] Figure 1is a schematic diagram of the computer device structure of the hardware operating environment described in the embodiment of the present invention;
[0055] Figure 2 is a flow chart of the industrial defect detection method according to an embodiment of the present invention;
[0056] Figure 3 It is a system block diagram of the industrial defect detection system described in an embodiment of the present invention. DETAILED DESCRIPTION
[0057] 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.
[0058] In order to solve the problem that existing industrial defect detection technology easily misses minor defects, this application provides an industrial defect detection method, system, device, medium and program product. Before introducing the specific technical solution of this application, the hardware operating environment involved in the embodiment of this application is first introduced.
[0059] See also Figure 1 , Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.
[0060] like Figure 1 As shown, the computer device may include: a processor, such as a central processing unit (CPU), a communication bus, a user interface, a network interface, and a memory. Among them, the communication bus is used to realize the connection and communication between these components. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and optionally, the user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity interface). The memory may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory may optionally be a storage device independent of the aforementioned processor.
[0061] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.
[0062] like Figure 1 As shown, the memory as a storage medium may include an operating system module, a network communication module, a user interface module and a data storage module.
[0063] exist Figure 1 In the computer device shown, the network interface is mainly used for data communication with the network server; the user interface is mainly used for data interaction with the user; the processor and memory in the computer device of the present application can be set in the computer device, and the computer device calls the computer program stored in the data storage module in the memory through the processor, and executes the industrial defect detection method provided in the embodiment of the present application.
[0064] Based on the hardware environment of the aforementioned embodiment, the embodiment of the present application provides an industrial defect detection method, please refer to Figure 2 , Figure 2 The flowchart of the industrial defect detection method is as follows:
[0065] (1) Perform image enlargement and image segmentation processing on the original image of the industrial product to obtain a region segmentation image.
[0066] In the embodiment of the present invention, the process of performing image enlargement processing and image segmentation processing on the original image of the industrial product is as follows:
[0067] A magnification factor of the original image is preset, and the original image is magnified based on the magnification factor, and the image sharpening and image denoising are performed on the original image after the image magnification to obtain the original magnified image; for example, the original image is first magnified four times, that is, the length and width of the original image are respectively magnified four times, and then the image sharpening is performed on the magnified original image, which can significantly enhance the edge and contour of the original image, making the details of the original image more prominent, and the image denoising is performed on the original image after the image sharpening, which can restore the detail information in the image, making the edges, textures and other features in the original image more distinct, thereby improving the recognition of the original image.
[0068] (1.2) A segmentation strategy for the original magnified image is preset, and image segmentation processing is performed on the original magnified image based on the segmentation strategy, and boundary optimization processing is performed on the original magnified image after image segmentation processing to obtain a region segmentation image. For example, if the magnification of the original image is four times, the magnified original image is evenly cut three times horizontally and vertically, and the magnified original image is cut to the original size, and a total of 16 region segmentation images with the same size as the original image are obtained.
[0069] (2) Perform image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image.
[0070] In an embodiment of the present invention, the process of performing image enhancement processing and edge contour extraction processing on the region segmentation image is as follows:
[0071] (2.1) Perform a forward Fourier transform on the region segmentation image to convert the region segmentation image from the time domain to the frequency domain.
[0072] (2.2) The region segmentation image in the frequency domain is subjected to high-frequency filtering and inverse Fourier transform to obtain an initial enhanced image. That is, since the edges and details of the region segmentation image are usually located in the high-frequency region, the edges and details of the region segmentation image can be enhanced by increasing the amplitude of the high-frequency component to achieve an enhancement effect.
[0073] (2.3) The Laplace operator is used to perform texture and edge detection processing on the initial enhanced image to obtain a regional contour image. Specifically, the template of the Laplace operator includes 4-neighborhood Laplace, 8-neighborhood Laplace, etc. The template can calculate the Laplace value of each pixel by performing a convolution operation with the image, and by setting a reasonable verification threshold, the pixels above or below the verification threshold are marked as edges or non-edges, thereby obtaining a regional contour image.
[0074] (3) Performing iterative contour enhancement processing on the regional contour image to obtain a contour enhanced image. The process of performing iterative contour enhancement processing on the regional contour image in the embodiment of the present invention is as follows:
[0075] (3.1) Preset adjustment parameters is a known positive number and is based on the adjustment parameter Calculate the conversion coefficient β of the region contour image.
[0076]
[0077] Among them, maxpixel represents the maximum grayscale value of the area contour image, It can be adjusted according to actual conditions.
[0078] (3.2) Based on the conversion coefficient, the grayscale value of the regional contour image is converted to the triangular space to obtain the image matrix corresponding to the regional contour image.
[0079] After1=Before*β
[0080] Among them, After1 represents the image matrix, and Before represents the grayscale value of the region contour image.
[0081] (3.3) Calculate the maximum value Max(After1) and the minimum value Min(After1) of the image matrix, and scale the image matrix to the grayscale space based on the maximum value and the minimum value of the image matrix.
[0082] After2=(After1-Min(After1))*(255 / Max(After1)-Min(After1))
[0083] Among them, After2 represents the grayscale value of the grayscale space.
[0084] (3.4) Preset the grayscale threshold, grayscale increase value, grayscale reduction value and number of iterative enhancements. For example, to set the grayscale threshold for iterative enhancement:
[0085] T=(Max(After2)-Min(After2)) / 2
[0086] Wherein, T represents the grayscale threshold, Max(After2) represents the maximum grayscale value of the grayscale space, and Min(After2) represents the minimum grayscale value of the grayscale space. Specifically, the grayscale increase value, grayscale reduction value, and iterative enhancement times are set according to actual conditions, and the grayscale increase value is generally smaller than the grayscale reduction value.
[0087] (3.5) Within the range of iterative enhancement times, the grayscale value of the regional contour image in the grayscale space is threshold-checked based on the grayscale threshold. If the grayscale value is greater than the grayscale threshold, the corresponding grayscale value is subjected to grayscale value difference processing according to the grayscale reduction value, and the remaining grayscale value is subjected to grayscale value addition processing according to the grayscale value increase value, so as to iteratively obtain the contour enhancement image of the industrial product. That is, through multiple iterative enhancements, the size of the grayscale value corresponding to the slight defect can be enhanced, thereby highlighting the small and slight defects, and the size of the grayscale value corresponding to the deep regular background image can be reduced, thereby weakening the deep and regular background.
[0088] (4) The contour enhanced image is processed by image stitching and image reduction to obtain the original enhanced image of the industrial product.
[0089] In the embodiment of the present invention, the process of performing image stitching processing and image reduction processing on the edge enhanced image is as follows:
[0090] (4.1) Perform image stitching on the contour enhanced images in reverse according to the segmentation strategy, and perform boundary fusion on the contour enhanced images after image stitching to obtain the original stitched image. That is, firstly perform image stitching on the contour enhanced images according to the positional relationship of the segmentation. If the magnification of the original image is four times, the size of the image after stitching is also four times of the original image; then perform boundary fusion on the stitched image to improve the image quality and enhance the visual effect.
[0091] (4.2) The original stitched image is reduced in reverse based on the magnification of the original image, and the image is sharpened and denoised on the reduced original stitched image to obtain the original enhanced image of the industrial product. That is, the original stitched image is reduced to the original image size, and the edge and contour of the original image can be significantly enhanced by sharpening the reduced original stitched image, so that the details of the original stitched image are more prominent, and the image denoising can be performed on the sharpened original stitched image to restore the detail information in the image, so that the edge, texture and other features in the original stitched image are more distinct, thereby improving the recognition of the original stitched image.
[0092] (5) The original image of the industrial product and the original enhanced image are associated and input into the defect detection system for defect detection to obtain defect location results and defect supplement location results.
[0093] In an embodiment of the present invention, the process of associating the original image of the industrial product and the original enhanced image and inputting them into the defect detection system for defect detection is as follows:
[0094] (5.1) The original image of the industrial product and the corresponding original enhanced image are correlated with each other.
[0095] (5.2) The mutually related original image and the original enhanced image are input into a pre-trained defect detection system (ADC) for detection, and the pre-trained defect detection system outputs a defect location result and a defect supplement location result respectively; wherein the defect location result and the defect supplement location result respectively include a defect position and a defect category.
[0096] (6) Comprehensively judge the defect location results and defect supplement location results to obtain the final defect detection results.
[0097] In the embodiment of the present invention, the process of comprehensively judging the defect location result and the defect supplement location result is as follows:
[0098] (6.1) Determine whether there is any overlap in defect locations based on the defect location results and defect supplement location results.
[0099] (6.2) If there is overlap in defect positions, continue to determine whether the corresponding defect types are the same. If the corresponding defect types are the same, retain the corresponding defect location results; if the defect types are different, retain the corresponding defect location results and defect supplementary location results to obtain the final defect detection results.
[0100] (6.3) If there is no overlap of defect positions, the corresponding defect location results and defect supplement location results are directly retained to obtain the final defect detection results.
[0101] In summary, the present invention can obtain more detailed texture features of the regional contour image by performing image magnification, image segmentation, image enhancement processing and edge contour extraction processing on the original image, and then perform contour iterative enhancement processing on the regional contour image, which can enhance the size of the grayscale value corresponding to the slight defect, thereby highlighting the small and slight defects, and reduce the size of the grayscale value corresponding to the deep regular background image, thereby weakening the deep and regular background; then perform image stitching and image reduction processing on the contour enhanced image, so that the size of the original enhanced image is the same as the original image size, and finally perform defect detection on the original image and the original enhanced image at the same time, which can effectively prevent the situation where slight defects are easily missed.
[0102] In addition, in one embodiment, based on the same inventive concept as the above embodiment, the embodiment of the present invention provides an industrial defect detection system, and the system corresponds to the method of embodiment 1 one by one, please refer to Figure 3 , Figure 3 The structural block diagram of the industrial defect detection system is as follows:
[0103] An image segmentation unit, wherein the original image segmentation unit is used to perform image magnification processing and image segmentation processing on the original image of the industrial product to obtain a region segmentation image;
[0104] A contour acquisition unit, the contour acquisition unit is used to perform image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image;
[0105] A contour enhancement unit, wherein the contour enhancement unit is used to perform contour iterative enhancement processing on the regional contour image to obtain a contour enhanced image;
[0106] An image stitching unit, the image stitching unit is used to perform image stitching processing and image reduction processing on the contour enhanced image to obtain an original enhanced image of the industrial product;
[0107] An associative positioning unit, the associative positioning unit is used to associate the original image and the original enhanced image of the industrial product and input them into the defect detection system for defect detection to obtain a defect positioning result and a defect supplementary positioning result;
[0108] The comprehensive determination unit is used to make a comprehensive determination on the defect location result and the defect supplement location result to obtain a final defect detection result.
[0109] It should be noted that each unit in the industrial defect detection system in this embodiment corresponds one-to-one to each step in the industrial defect detection method in the aforementioned embodiment. Therefore, the specific implementation method and technical effects achieved in this embodiment can refer to the implementation method of the aforementioned industrial defect detection method and will not be repeated here.
[0110] In addition, in one embodiment, the present application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory, and the computer program implements the method in the aforementioned embodiment when executed by the processor.
[0111] In addition, in one embodiment, the present application further provides a computer storage medium, on which a computer program is stored, and the computer program implements the method in the aforementioned embodiment when executed by a processor.
[0112] In some embodiments, the computer readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or various devices including one or any combination of the above memories. The computer may be various computing devices including intelligent terminals and servers.
[0113] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.
[0114] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).
[0115] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.
[0116] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.
[0117] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0119] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. An industrial defect detection method, characterized in that: The method comprises the following steps: Perform image magnification and image segmentation processing on the original image of the industrial product to obtain a region segmentation image; Performing image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image; Performing iterative contour enhancement processing on the regional contour image to obtain a contour enhanced image; Perform image stitching and image reduction processing on the contour enhanced image to obtain the original enhanced image of the industrial product; The original image and the original enhanced image of the industrial product are associated and input into the defect detection system for defect detection to obtain defect location results and defect supplement location results; The defect location results and defect supplement location results are comprehensively judged to obtain the final defect detection results.
2. An industrial defect detection method according to claim 1, characterized in that: The process of image enlargement and image segmentation of the original image of industrial products is as follows: Presetting a magnification of the original image, performing image magnification processing on the original image based on the magnification factor, and performing image sharpening processing and image denoising processing on the original image after the image magnification processing to obtain the original magnified image; A segmentation strategy of the original enlarged image is preset, and image segmentation processing is performed on the original enlarged image based on the segmentation strategy, and boundary optimization processing is performed on the original enlarged image after image segmentation processing to obtain a region segmentation image.
3. The industrial defect detection method according to claim 1, characterized in that: The process of image enhancement and edge contour extraction for the region segmentation image is as follows: Performing forward Fourier transform on the region segmentation image to convert the region segmentation image from the time domain to the frequency domain; Performing high-frequency filtering and inverse Fourier transform on the regional segmented image in the frequency domain to obtain an initial enhanced image; The Laplacian operator is used to perform texture and edge detection on the initial enhanced image to obtain a region contour image.
4. The industrial defect detection method according to claim 1, characterized in that: The process of iterative contour enhancement processing on the regional contour image is as follows: Presetting adjustment parameters, and calculating conversion coefficients of the region contour image based on the adjustment parameters; The grayscale value of the regional contour image is converted into a triangular space based on the conversion coefficient to obtain an image matrix corresponding to the regional contour image; Calculate the maximum and minimum values of the image matrix, and scale the image matrix to the grayscale space based on the maximum and minimum values of the image matrix; Preset the grayscale threshold, grayscale increase value, grayscale reduction value and number of iterative enhancements for iterative enhancement; Within the range of iterative enhancement times, the grayscale value of the regional contour image in the grayscale space is threshold checked based on the grayscale threshold. If the grayscale value is greater than the grayscale threshold, the grayscale value is subtracted according to the grayscale reduction value, and the remaining grayscale value is added according to the grayscale increase value, so as to iteratively obtain the contour enhanced image of the industrial product.
5. The industrial defect detection method according to claim 1, characterized in that: The process of image stitching and image reduction processing for contour enhanced images is as follows: Performing image stitching processing on the contour enhanced image in reverse according to the segmentation strategy, and performing boundary fusion on the contour enhanced image processed by image stitching to obtain the original stitching image; The original stitched image is reversely reduced based on the magnification of the original image, and the image is sharpened and denoised to obtain the original enhanced image of the industrial product.
6. An industrial defect detection method according to claim 1, characterized in that: The process of associating the original image of the industrial product and the original enhanced image and inputting them into the defect detection system for defect detection is as follows: Correlating an original image of an industrial product and a corresponding original enhanced image; The mutually related original image and the original enhanced image are input into a pre-trained defect detection system for detection, and the pre-trained defect detection system outputs a defect location result and a defect supplement location result respectively; wherein the defect location result and the defect supplement location result respectively include a defect position and a defect category.
7. An industrial defect detection method according to claim 6, characterized in that: The process of comprehensively judging the defect location results and defect supplement location results is as follows: Determine whether there is overlap in defect positions based on the defect location results and defect supplement location results; If there is overlap in defect positions, continue to determine whether the corresponding defect types are the same. If the corresponding defect types are the same, retain the corresponding defect location results; If the defect types are different, the corresponding defect location results and defect supplement location results are retained to obtain the final defect detection result; If there is no overlap in defect positions, the corresponding defect location results and defect supplement location results are directly retained to obtain the final defect detection result.
8. An industrial defect detection system, characterized in that: The system comprises: An image segmentation unit, wherein the original image segmentation unit is used to perform image magnification processing and image segmentation processing on the original image of the industrial product to obtain a region segmentation image; A contour acquisition unit, the contour acquisition unit is used to perform image enhancement processing and edge contour extraction processing on the region segmentation image to obtain a region contour image; A contour enhancement unit, wherein the contour enhancement unit is used to perform contour iterative enhancement processing on the regional contour image to obtain a contour enhanced image; An image stitching unit, the image stitching unit is used to perform image stitching processing and image reduction processing on the contour enhanced image to obtain an original enhanced image of the industrial product; An associative positioning unit, the associative positioning unit is used to associate the original image and the original enhanced image of the industrial product and input them into the defect detection system for defect detection to obtain a defect positioning result and a defect supplementary positioning result; The comprehensive determination unit is used to make a comprehensive determination on the defect location result and the defect supplement location result to obtain a final defect detection result.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the industrial defect detection method as described in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the industrial defect detection method according to any one of claims 1 to 7 is implemented.
11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, an industrial defect detection method as described in any one of claims 1 to 7 is implemented.