A method, device, electronic device and storage medium for laser cleaning of paint rust selected areas

Abstract

The present invention provides a method, device, electronic device and storage medium for selecting areas for laser cleaning of paint and rust, and relates to the technical field of laser cleaning. The method comprises S1, based on a standard light source, performing image sampling on a workpiece to be cleaned to generate an RGB sampling image; S2, performing coordinate calibration on the RGB sampling image to respectively generate calibration graphic size and coordinate reference; S3, performing texture feature analysis on the RGB sampling image according to the calibration graphic size and the coordinate reference to generate a rust partitioning result; S4, based on the LAB color space, using a color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set; S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result. The present invention can improve the quality of laser cleaning.

Description

Technical Field

[0001] The present invention relates to the technical field of laser cleaning, and in particular to a method, device, electronic equipment and storage medium for laser cleaning of paint and rust selected areas. Background Art

[0002] Laser cleaning technology is an important emerging branch of laser surface processing technology. It plays an important role in quality maintenance and performance assurance in the field of modern industrial manufacturing. Laser cleaning technology is a green and non-destructive surface cleaning method that uses the interaction between laser and matter to effectively remove attachments on the surface of the substrate. It can also use different removal principles such as thermal ablation, resonance, and thermal expansion to jointly realize the cleaning process of the workpiece surface.

[0003] In order to achieve an efficient, high-quality laser cleaning process that is highly integrated with intelligent control, the existing laser cleaning technology includes steps such as graphic recognition of cleaning targets, coverage path planning, and dynamic control of vector parameters such as laser incident angle, scanning rate and scanning mode. Although these steps can improve the laser cleaning effect to a certain extent, due to the different processing conditions of the cleaning targets, there are complex paint, rust and dirt complex situations, and the current image analysis method used for laser cleaning is relatively simple, and it is impossible to make effective extraction of the area to be cleaned for complex cleaning scenarios, which affects the final quality of laser cleaning. Summary of the invention

[0004] The problem solved by the present invention is how to effectively extract the to-be-cleaned area of ​​the cleaning target to improve the final quality of laser cleaning.

[0005] In a first aspect, the present invention provides a method for laser cleaning of paint rust selected areas, comprising:

[0006] S1, based on the standard light source, performs image sampling on the workpiece to be cleaned and generates an RGB sampling image;

[0007] S2, calibrating the coordinates of the RGB sampling image to generate a calibration graphic size and a coordinate reference respectively;

[0008] S3, performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result;

[0009] S4, based on the LAB color space, using the color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set;

[0010] S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result.

[0011] Optionally, the performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result includes:

[0012] According to the calibration figure size and the coordinate reference, a local entropy value filtering method is used to divide the texture features of the RGB sampling image to generate a texture image;

[0013] The texture image is segmented according to the RGB sampling image to generate the rust partition result.

[0014] Optionally, segmenting the texture image according to the RGB sampling image to generate the rust partition result includes:

[0015] Extracting and reconstructing grayscale of the texture image to generate a grayscale image;

[0016] According to the grayscale image, mapping the RGB sampling image to generate a texture feature image;

[0017] The texture feature image is segmented according to the RGB sampling image to generate the rust partition result.

[0018] Optionally, the paint color segmentation is performed on the RGB sampling image based on the LAB color space by using a color segmentation method to generate a paint color coordinate distribution set, including:

[0019] Based on the LAB color space, converting the RGB sampled image into a LAB pattern;

[0020] Extracting the Euclidean distance between each pixel in the LAB pattern and the K-color standard color card, and generating a color index order according to the Euclidean distance;

[0021] The paint color coordinate distribution set is generated according to the color index order and the RGB sampling image.

[0022] Optionally, performing failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result includes:

[0023] Performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values;

[0024] Reusing the rust partition result as a taboo area;

[0025] According to the taboo area, the paint color coordinate distribution set is invalidated and mapped, and according to the color pixel value, the paint color segmentation result is generated.

[0026] Optionally, performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values ​​includes:

[0027] According to the paint color coordinate distribution set, an image matrix formula group is used to perform background filling on the RGB sampling image to generate the color pixel value, and the image matrix formula group includes:

[0028]

[0029] in, is the image matrix of monochrome classification in the paint color coordinate distribution set, f p,q is the color pixel value of the RGB sampled image at point (p, q), C U A is the complement of the matrix coordinates of the RGB sampled image, C k is the paint color coordinate distribution set, kth color classification is not black means that the result of the kth paint color segmentation is not black, and kth color classification is black means that the result of the kth paint color segmentation is black.

[0030] Optionally, after performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result and generating the paint color segmentation result, the method further includes:

[0031] According to the cleaning progress, steps S1-S5 are repeated until the cleaning is completed.

[0032] In a second aspect, the present invention provides a laser cleaning device for selecting a paint rust area, comprising:

[0033] A sampling module is used to sample the image of the workpiece to be cleaned based on a standard light source and generate an RGB sampling image;

[0034] A coordinate module, used for calibrating the coordinates of the RGB sampling image, and generating a calibration graphic size and a coordinate reference respectively;

[0035] A texture module, used to perform texture feature analysis on the RGB sampling image according to the calibration graphic size and the coordinate reference, and generate a rust partition result;

[0036] A paint color module is used to perform paint color segmentation on the RGB sampling image based on the LAB color space and adopt a color segmentation method to generate a paint color coordinate distribution set;

[0037] A mapping module is used to perform failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result. A generating module is used to perform failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result.

[0038] In a third aspect, the present invention provides an electronic device, including a memory and a processor;

[0039] The memory is used to store computer programs;

[0040] The processor is used to implement the laser cleaning paint rust selection method as described in the first aspect when executing the computer program.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for laser cleaning of paint and rust selected areas as described in the first aspect is implemented.

[0042] The beneficial effects of the laser cleaning paint rust selection method, device, electronic equipment and storage medium of the present invention are:

[0043] Auxiliary illumination by standard light source can make the workpiece to be cleaned be sampled under uniform light illumination, and obtain standard RGB sampling image, which is convenient for subsequent texture feature analysis and paint color segmentation, and indirectly improves the efficiency of extracting the area to be cleaned of the cleaning target. Then, the coordinates of the RGB sampling image are calibrated, and the calibration graphic size and coordinate reference are generated respectively. According to the calibration graphic size and coordinate reference, the texture feature analysis of the RGB sampling image is performed, and the randomness and texture distribution characteristics of the dirt on the paint-rust composite surface are comprehensively considered, the image texture distribution difference is analyzed, the rust partition is accurately located, and the accurate rust partition result is obtained. At the same time, based on the LAB color space, it can better simulate The color space perceived by the human eye can therefore more accurately represent the color in the image. The RGB sampling image is then segmented by the color segmentation method, which can separate the pixels of different colors in the image and generate a paint color coordinate distribution set, that is, record the position and distribution of different colors in the image to obtain a paint color coordinate distribution set. Finally, according to the rust zoning result, the paint color coordinate distribution set is invalidated and mapped to cope with the complex situation of rusty paint including black paint or white paint, and the paint color segmentation result is obtained. According to the rust zoning result and the paint color segmentation result, the accurate area to be cleaned can be determined, which is convenient for subsequent laser cleaning to improve the final quality of laser cleaning. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a process for laser cleaning of paint rust selected areas provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a color space provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the structure of the intelligent area selection system for laser cleaning provided by an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of the structure of a laser paint rust selection cleaning device provided in an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of the corrosion partitioning result provided by an embodiment of the present invention;

[0050] Figure 7 A schematic diagram of a paint color segmentation result provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0052] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0053] The term "including" and its variations used in this article are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units. It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0054] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0055] In view of the problems existing in the above-mentioned related technologies, this embodiment provides a method, device, electronic device and storage medium for laser cleaning of paint and rust selected areas.

[0056] like Figure 1 As shown, a method for laser cleaning of paint rust selected areas provided by an embodiment of the present invention includes:

[0057] S1, based on the standard light source, performs image sampling on the workpiece to be cleaned and generates an RGB sampling image.

[0058] Specifically, a standard light source is used to perform auxiliary lighting on the image sampling scene of the workpiece to be cleaned, and then the workpiece to be cleaned is sampled to obtain an RGB sampling image, and the RGB sampling image is compressed into an image data output package in a readable form for subsequent steps. For example, a standard light source such as a D65 standard fill light can be used to perform auxiliary lighting on the image sampling scene of the workpiece to be cleaned, and an image acquisition device such as a CCD industrial camera can also be used to perform image sampling on the workpiece to be cleaned. When the CCD industrial camera is used for image sampling, the workpiece to be cleaned is first adjusted and fixed according to the mechanical coordinate reference. The mechanical coordinate reference reference is used to adjust and match the coordinate set of the target area, and then the cross section of the workpiece to be cleaned is orthogonal to the visual axis of the CCD industrial camera, and it is ensured that it is completely presented in the field of view of the image acquisition module, and then the image sampling scene of the workpiece to be cleaned is performed using a standard light source. Finally, the CCD industrial camera is used to perform image sampling on the workpiece to be cleaned to generate an RGB sampling image.

[0059] S2, calibrating the coordinates of the RGB sampling image to generate a calibration graphic size and a coordinate reference respectively.

[0060] Specifically, the coordinates of the RGB sampling image are calibrated to generate the calibration graphic size and coordinate reference respectively, so as to determine the image position and thus ensure the accuracy and precision of the selected area.

[0061] S3, performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result.

[0062] Specifically, after obtaining the calibration graphic size and coordinate reference and determining the image position, the grayscale co-occurrence matrix, texture direction histogram, local binary pattern, directional gradient histogram, and local entropy filtering methods can be used to perform texture feature analysis on the RGB sampling image, analyze the texture distribution differences of the image, and obtain the rust partitioning results, so as to determine the rusted area on the workpiece to be cleaned, and obtain the rust partitioning results, wherein the rust partitioning results include the spatial coordinate set of the effective classification features in each sub-area obtained by the texture feature analysis, that is, the effective coordinates corresponding to the texture sub-area are obtained. Exemplarily, the rust partitioning results are as follows: Figure 6 As shown, the white area is the characteristic rust area, and the black area is the overall paint distribution area.

[0063] S4, based on the LAB color space, using the color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set.

[0064] Specifically, step S4 can be performed while step S3 is being performed, that is, S3 and S4 can be performed in parallel, so as to speed up the operation rate and increase the processing efficiency. The LAB color space is a color model defined by the International Commission on Illumination (abbreviated as CIE) in 1976. The color space schematic diagram is as follows: Figure 2 As shown, the LAB color model uses three channels to represent colors: the "L" channel is responsible for perceived brightness, the "a" channel and the "b" channel are responsible for color distribution, where the "a" channel represents the positive and negative range from red to green, and the "b" channel represents the positive and negative range from yellow to blue. Since the color gamut of the LAB color space is larger than that of the RGB space and the CMYK space, and it is a color expression mode that digitally fits human visual perception and is independent of the device, it can make up for the deficiency that the RGB and CMYK modes must rely on the color characteristics of the device. Therefore, based on the color resolution ability of the LAB color space, the color segmentation method is used to perform color-based recognition and segmentation on the sampling pattern, and the coordinate distribution set of the RGB sampling image for different colors, that is, the paint color coordinate distribution set, can be obtained.

[0065] S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result.

[0066] Specifically, after obtaining the rust zoning result, the rust zoning result is reused as the taboo interval, and the paint color coordinate distribution set is invalidated and mapped according to the taboo interval, that is, background mapping, and the paint color coordinate distribution set is subtracted to distinguish each paint color from the rust area, and the final coordinate position corresponding to each paint color is obtained, that is, the paint color segmentation result. According to the rust zoning result and the paint color segmentation result, the positions of the rusted areas and different paint colors on the workpiece to be cleaned can be determined, so as to determine the area to be processed by laser cleaning, that is, the area to be cleaned, according to the actual cleaning requirements, so as to facilitate the next step of laser cleaning.

[0067] In this embodiment, auxiliary lighting is performed by a standard light source, so that the workpiece to be cleaned can be sampled under uniform light illumination to obtain a standard RGB sampling image, which is convenient for subsequent texture feature analysis and paint color segmentation, and indirectly improves the efficiency of extracting the area to be cleaned of the cleaning target. Then, an image acquisition device such as a CCD industrial camera is used to calibrate the coordinates of the RGB sampling image, and a calibration graphic size and a coordinate reference are generated respectively. According to the calibration graphic size and the coordinate reference, a texture feature analysis is performed on the RGB sampling image. The randomness and texture distribution characteristics of the dirt on the paint-rust composite surface are comprehensively considered, the image texture distribution difference is analyzed, the rust partition is accurately located, and an accurate rust partition result is obtained. At the same time, based on the LAB color space, it can better The color space perceived by the human eye can be simulated accurately, so the color in the image can be represented more accurately. The RGB sampling image is then segmented by paint color through the color segmentation method. The pixels of different colors in the image can be segmented and a paint color coordinate distribution set is generated, that is, the position and distribution of different colors in the image are recorded to obtain a paint color coordinate distribution set. Finally, according to the rust zoning result, the paint color coordinate distribution set is invalidated and mapped to cope with the complex situation of rusty paint including black paint or white paint, and the paint color segmentation result is obtained. According to the rust zoning result and the paint color segmentation result, the accurate area to be cleaned is obtained, which is convenient for providing accurate selection basis for the classification and design of subsequent laser cleaning process parameters to improve the final quality of laser cleaning.

[0068] In one embodiment, the laser cleaning paint rust selection method can be applied to a laser cleaning intelligent selection system, such as Figure 3As shown, the intelligent area selection system for laser cleaning includes a workpiece to be cleaned, a standard light source, a graphics acquisition and processing module coaxially assembled with the cleaning laser, and an intelligent identification and selection module, wherein the graphics acquisition and processing module is used to perform image sampling on the workpiece to be cleaned based on the standard light source, generate an RGB sampling image, and compress the RGB sampling image into a corresponding image data output package, and upload the RGB sampling image to the intelligent identification and selection module in a readable form; the intelligent identification and selection module is used to read the RGB sampling image, and perform coordinate calibration on the RGB sampling image, and respectively generate a calibration graphic size and a coordinate reference; according to the calibration graphic size and the coordinate reference, the RGB sampling image is subjected to texture feature analysis to generate a rust partitioning result; based on the LAB color space, the color segmentation method is adopted to perform paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set; according to the rust partitioning result, the paint color coordinate distribution set is subjected to failure mapping to generate a paint color segmentation result.

[0069] Optionally, the performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result includes:

[0070] According to the calibration figure size and the coordinate reference, a local entropy value filtering method is used to divide the texture features of the RGB sampling image to generate a texture image;

[0071] The texture image is segmented according to the RGB sampling image to generate the rust partition result.

[0072] Specifically, the local entropy filtering method can comprehensively consider the randomness and texture distribution characteristics of the paint-rust composite surface dirt, and take into account the operation cycle and the complexity of the operation. The texture differences of the paint-rust are clearly divided by using the local entropy texture filtering method. The obtained texture image can accurately segment the RGB sampling image and generate the rust partition result.

[0073] Exemplarily, the adopting of a local entropy value filtering method to divide the texture features of the RGB sampling image to generate a texture image includes:

[0074] The entropy value of the local subdomain is determined according to the local entropy formula, which includes:

[0075]

[0076] Where H(u,v) is the local subdomain entropy value at point (u,v), and 0≤u≤N-M+1, 0≤v≤N-M+1, where N is the current RGB sampled image (F u,v ), that is, the size of the current RGB sampling image (F u,v) is N×N, M is the size of a local subdomain determined according to the reference point (u, v), that is, the scale of the local subdomain is M×M, f u+i,v+j is the pixel value in the subdomain, and i, j are both recorded as the current value range of the subdomain, the range relationship is 0≤i≤M-1, 0≤j≤M-1, P i,j (u,v) is the probability at the reference point (u,v), and the probability formula includes:

[0077]

[0078] All pixels in the RGB sampled image are traversed sequentially to obtain the entropy values ​​of all local subdomains corresponding to the complete image, and then a texture image containing all randomness statistical measures is obtained. For example, the scale of the local subdomain can be set to 9×9.

[0079] Optionally, segmenting the texture image according to the RGB sampling image to generate the rust partition result includes:

[0080] Extracting and reconstructing grayscale of the texture image to generate a grayscale image;

[0081] According to the grayscale image, mapping the RGB sampling image to generate a texture feature image;

[0082] The texture feature image is segmented according to the RGB sampling image to generate the rust partition result.

[0083] Specifically, the entropy information contained in the texture image after statistical processing reflects the disorder of pixel composition and distribution. The texture image is extracted and reconstructed in grayscale to generate a grayscale image. The texture feature image of the sampling image is mapped with the change of grayscale distribution to generate a texture feature image. At this time, the pixel distribution difference on the surface of the RGB sampling image is referred to, and the texture areas with different gradients are segmented to obtain the final rust partitioning result. Exemplarily, according to the different grayscale distribution characteristics of the paint surface and the rust after local entropy processing, the number of grayscale segments is 2, and the segmentation interval is [0, 80%] and (80%, 100%], that is, the grayscale threshold of the texture filtering and segmentation of the rust is 0.8. At the same time, the binary result obtained by the above texture filtering is expanded and opened and closed with appropriate size to obtain the binary segmentation result of the continuous area of ​​the paint rust texture feature, that is, the rust partitioning result.

[0084] Optionally, the paint color segmentation is performed on the RGB sampling image based on the LAB color space by using a color segmentation method to generate a paint color coordinate distribution set, including:

[0085] Based on the LAB color space, converting the RGB sampled image into a LAB pattern;

[0086] Extracting the Euclidean distance between each pixel in the LAB pattern and the K-color standard color card, and generating a color index order according to the Euclidean distance;

[0087] The paint color coordinate distribution set is generated according to the color index order and the RGB sampling image.

[0088] Specifically, the RGB sampled image is converted into a LAB pattern through a conversion formula group, and the conversion formula group includes:

[0089]

[0090] Among them, X, Y and Z are the value variables of each channel in the XYZ color space, and R, G and B are the value variables of each channel in the RGB color space.

[0091] Then there are expressions for each channel in LAB space:

[0092] L=116f(Y / Y n )-16;

[0093] a=500[f(X / X n )-f(Y / Y n )];

[0094] b=200[f(Y / Y n )-f(Z / Z n )];

[0095] Among them, X n , Y n , Z n is the mapping coefficient of the color space, L, a and b are the value variables representing each channel of the LAB color space, and the function f with t as the independent variable is expressed as:

[0096]

[0097] After the RGB sampled image is converted into a LAB pattern, the LAB pattern is color-resolved and measured with reference to the LAB color chart under the standard light source. The image under the LAB model is simplified as a two-dimensional array. The color of each pixel represented by the two-dimensional coordinates in the image is obtained by coupling the L, a, and b channel measurements. If the current LAB image (G r,s) has a scale of N×N, then the two-dimensional coordinate index range of the pixel point c(r,s) is 0≤r≤N-1, 0≤s≤N-1. At this time, the Euclidean distance of each pixel point from the K-color standard color card is traversed, and the color order corresponding to the minimum Euclidean distance calculated for each pixel point is the color index order that best matches the pixel point. Finally, the color index order and the RGB sampling image are sorted to generate the paint color coordinate distribution set.

[0098] The step of extracting the Euclidean distance between each pixel in the LAB pattern and the K color standard color card includes:

[0099] The Euclidean distance formula is used to extract the Euclidean distance between each pixel in the LAB pattern and the K color standard color card. The Euclidean distance formula includes:

[0100]

[0101] Where, ΔE c (k) is the Euclidean distance, k is the kth order, L c is the L channel variable value traversed to the current pixel position, L k is the L channel variable value of the current standard color card color, a c is the a channel variable value traversed to the current pixel position, a k is the a channel variable value under the current order standard color card color, b c is the b channel variable value traversed to the current pixel position, b k is the b channel variable value under the current order standard color card color, and K means there are K standard color card orders for comparison.

[0102] Optionally, performing failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result includes:

[0103] Performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values;

[0104] Reusing the rust partition result as a taboo area;

[0105] According to the taboo area, the paint color coordinate distribution set is invalidated and mapped, and according to the color pixel value, the paint color segmentation result is generated.

[0106] In one embodiment, the background filling of the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values ​​includes:

[0107] According to the paint color coordinate distribution set, an image matrix formula group is used to perform background filling on the RGB sampling image to generate the color pixel value, and the image matrix formula group includes:

[0108]

[0109] in, is the image matrix of monochrome classification in the paint color coordinate distribution set, f p,q is the color pixel value of the RGB sampled image at point (p, q), C U A is the complement of the matrix coordinates of the RGB sampled image, C k is the paint color coordinate distribution set, kth color classification is not black means that the result of the kth paint color segmentation is not black, and kth color classification is black means that the result of the kth paint color segmentation is black.

[0110] Specifically, when obtaining the color pixel value, The invalid processing of the rust area is included, and the taboo coordinate set C obtained by the paint rust texture segmentation process is b Cross-multiplexing, in each monochrome partition graphic matrix In the process, the mapping formula is used to carry out the taboo area mapping process, and the mapping formula includes:

[0111]

[0112] in, is the image matrix of monochrome classification in the paint color coordinate distribution set updated after taboo area mapping, C b When the paint includes black or white, according to the above formula, when the paint includes black, the background is set to white, such as Figure 7 As shown, the background of black paint is set to white. When the paint includes white, the background is set to black. Update the pixel value and finally obtain the paint color segmentation result.

[0113] Optionally, after performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result and generating the paint color segmentation result, the method further includes:

[0114] According to the cleaning progress, steps S1-S5 are repeated until the cleaning is completed.

[0115] Specifically, before and after the laser cleaning process, steps S1-S5 are performed at a certain frequency in each cleaning stage of the workpiece surface to be cleaned, so as to monitor the cleaning effect in real time, determine the degree and condition of the rusted area, and then facilitate the formulation of cleaning strategies to improve the cleaning effect.

[0116] like Figure 4 As shown, a laser cleaning paint rust selection device 400 provided in an embodiment of the present invention includes:

[0117] The sampling module 410 is used to perform image sampling on the workpiece to be cleaned based on a standard light source and generate an RGB sampling image;

[0118] A coordinate module 420 is used to calibrate the coordinates of the RGB sampled image and generate a calibration graphic size and a coordinate reference respectively;

[0119] A texture module 430 is used to perform texture feature analysis on the RGB sampling image according to the calibration graphic size and the coordinate reference to generate a rust partition result;

[0120] A paint color module 440 is used to perform paint color segmentation on the RGB sampling image based on the LAB color space and adopt a color segmentation method to generate a paint color coordinate distribution set;

[0121] The mapping module 450 is used to perform failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result.

[0122] like Figure 5 As shown, an electronic device 500 provided by an embodiment of the present invention includes a memory 510 and a processor 520; the memory 510 is used to store a computer program; the processor 520 is used to implement the above-mentioned laser paint rust cleaning area selection method when executing the computer program.

[0123] In other words, an electronic device 500 includes a memory 510 and a processor 520 coupled to the memory 510; the memory 510 is configured to store a computer program; and the processor 520 is configured to perform the following operations when executing the computer program:

[0124] S1, based on the standard light source, performs image sampling on the workpiece to be cleaned and generates an RGB sampling image;

[0125] S2, calibrating the coordinates of the RGB sampling image to generate a calibration graphic size and a coordinate reference respectively;

[0126] S3, performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result;

[0127] S4, based on the LAB color space, using the color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set;

[0128] S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result.

[0129] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned laser cleaning method for selecting paint and rust areas is implemented.

[0130] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations:

[0131] S1, based on the standard light source, performs image sampling on the workpiece to be cleaned and generates an RGB sampling image;

[0132] S2, calibrating the coordinates of the RGB sampling image to generate a calibration graphic size and a coordinate reference respectively;

[0133] S3, performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result;

[0134] S4, based on the LAB color space, using the color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set;

[0135] S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result.

[0136] An electronic device 500 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 500 is intended to represent various forms of digital electronic computer equipment, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 500 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present invention described and / or required herein.

[0137] The electronic device 500 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0138] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0139] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A laser cleaning method for selected areas of paint rust, characterized in that: include: S1, based on the standard light source, performs image sampling on the workpiece to be cleaned and generates an RGB sampling image; S2, calibrating the coordinates of the RGB sampling image to generate a calibration graphic size and a coordinate reference respectively; S3, performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result; S4, based on the LAB color space, using the color segmentation method, performing paint color segmentation on the RGB sampling image to generate a paint color coordinate distribution set; S5, performing failure mapping on the paint color coordinate distribution set according to the rust partitioning result to generate a paint color segmentation result; The paint color segmentation is performed on the RGB sampling image based on the LAB color space by using the color segmentation method to generate a paint color coordinate distribution set, including: Based on the LAB color space, converting the RGB sampled image into a LAB pattern; Extracting the Euclidean distance between each pixel in the LAB pattern and the K-color standard color card, and generating a color index order according to the Euclidean distance, wherein the color order corresponding to the minimum Euclidean distance calculated for each pixel is the color index order that best matches the pixel; Generating the paint color coordinate distribution set according to the color index order and the RGB sampling image; The step of performing failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result includes: Performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values; Reusing the rust partition result as a taboo area; According to the taboo area, failure mapping is performed on the paint color coordinate distribution set, and according to the color pixel value, the paint color segmentation result is generated; The step of performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values ​​includes: According to the paint color coordinate distribution set, an image matrix formula group is used to perform background filling on the RGB sampling image to generate the color pixel value, and the image matrix formula group includes: in, is the graphic matrix of the monochrome classification in the paint color coordinate distribution set, f p,q is the color pixel value of the RGB sampled image at point (p, q), C U A is the complement of the matrix coordinates of the RGB sampled image, C k is the paint color coordinate distribution set; The taboo coordinate set C obtained from the paint rust texture segmentation process b Cross-multiplexing, the graphic matrix in each monochrome partition In the process, a mapping formula is used to perform taboo area mapping processing, and the mapping formula includes: in, is the image matrix of monochrome classification in the paint color coordinate distribution set updated after taboo area mapping, C b It is a forbidden area.

2. The laser cleaning method for selected areas of paint rust according to claim 1, characterized in that: The step of performing texture feature analysis on the RGB sampling image according to the calibrated graphic size and the coordinate reference to generate a rust partition result includes: According to the calibration figure size and the coordinate reference, a local entropy value filtering method is used to divide the texture features of the RGB sampling image to generate a texture image; The texture image is segmented according to the RGB sampling image to generate the rust partition result.

3. The laser cleaning method for selected areas of paint rust according to claim 2 is characterized in that: The step of segmenting the texture image according to the RGB sampling image and generating the rust partition result includes: Extracting and reconstructing grayscale of the texture image to generate a grayscale image; According to the grayscale image, mapping the RGB sampling image to generate a texture feature image; The texture feature image is segmented according to the RGB sampling image to generate the rust partition result.

4. The laser cleaning method for selected areas of paint rust according to any one of claims 1 to 3, characterized in that: According to the rust partitioning result, after performing failure mapping on the paint color coordinate distribution set and generating the paint color segmentation result, the method further includes: According to the cleaning progress, steps S1-S5 are repeated until the cleaning is completed.

5. A laser cleaning device for selecting paint and rust areas, characterized in that: include: A sampling module is used to sample the image of the workpiece to be cleaned based on a standard light source and generate an RGB sampling image; A coordinate module, used for calibrating the coordinates of the RGB sampling image, and generating a calibration graphic size and a coordinate reference respectively; A texture module, used to perform texture feature analysis on the RGB sampling image according to the calibration graphic size and the coordinate reference, and generate a rust partition result; A paint color module is used to perform paint color segmentation on the RGB sampling image based on the LAB color space and adopt a color segmentation method to generate a paint color coordinate distribution set; A mapping module, used to perform failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result; The paint color segmentation is performed on the RGB sampling image based on the LAB color space by using the color segmentation method to generate a paint color coordinate distribution set, including: Based on the LAB color space, converting the RGB sampled image into a LAB pattern; Extracting the Euclidean distance between each pixel in the LAB pattern and the K-color standard color card, and generating a color index order according to the Euclidean distance, wherein the color order corresponding to the minimum Euclidean distance calculated for each pixel is the color index order that best matches the pixel; Generating the paint color coordinate distribution set according to the color index order and the RGB sampling image; The step of performing failure mapping on the paint color coordinate distribution set according to the rust partition result to generate a paint color segmentation result includes: Performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values; Reusing the rust partition result as a taboo area; According to the taboo area, failure mapping is performed on the paint color coordinate distribution set, and according to the color pixel value, the paint color segmentation result is generated; The step of performing background filling on the RGB sampling image according to the paint color coordinate distribution set to generate color pixel values ​​includes: According to the paint color coordinate distribution set, an image matrix formula group is used to perform background filling on the RGB sampling image to generate the color pixel value, and the image matrix formula group includes: in, is the graphic matrix of the monochrome classification in the paint color coordinate distribution set, f p,q is the color pixel value of the RGB sampled image at point (p, q), C U A is the complement of the matrix coordinates of the RGB sampled image, C k is the paint color coordinate distribution set; The taboo coordinate set C obtained from the paint rust texture segmentation process b Cross-multiplexing, the graphic matrix in each monochrome partition In the process, a mapping formula is used to perform taboo area mapping processing, and the mapping formula includes: in, is the image matrix of monochrome classification in the paint color coordinate distribution set updated after taboo area mapping, C b It is a forbidden area.

6. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the laser paint rust selection cleaning method as described in any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the method for laser cleaning of paint and rust selected areas as described in any one of claims 1 to 4 is implemented.

Images