Method for detecting and optimizing appearance of plating layer of electroplating solution

Abstract

The invention discloses a plating appearance detection optimization method of an electroplating solution, and relates to the related technical field of plating detection optimization, and the method comprises the following steps: carrying out appearance quality detection on a plating sample; if the color deviation coefficient is greater than or equal to a first deviation coefficient threshold value, or / and the thickness deviation coefficient is greater than or equal to a second deviation coefficient threshold value, or / and the roughness deviation coefficient is greater than or equal to a third deviation coefficient threshold value, obtaining electroplating control parameters of the plating sample; constructing a fitness evaluation function according to the deviation index; performing optimization according to the fitness evaluation function; and carrying out plating appearance optimization according to an optimization result. The technical problem that product quality fluctuates and is different due to the fact that existing electroplating solution plating layer detection depends on experience to conduct plating layer appearance detection and is high in subjectivity, low in efficiency and poor in consistency is solved, accurate, efficient and automatic plating layer appearance detection is achieved, and the technical effect of improving product quality stability and consistency is achieved.

Description

Technical Field

[0001] The present application relates to the field related to coating detection optimization technology, and specifically to a coating appearance detection optimization method for an electroplating solution. Background Art

[0002] With the rapid development of the electronics industry and the automobile industry, the requirements for the appearance quality of the coating of the electroplating solution are becoming higher and higher. It is necessary to build an intelligent coating appearance detection system to realize fast, accurate and automatic detection of the coating appearance of the electroplating solution. The optimization of the coating appearance detection of the electroplating solution has become an important issue that needs to be urgently solved in the current manufacturing industry. The traditional appearance detection based on experience cannot meet the needs of accurate evaluation and efficient detection of the coating appearance. The traditional coating appearance detection is easily affected by human subjective factors, resulting in inconsistent judgment results and frequent misjudgments, which seriously affects the quality stability and consistency of the product. In addition, the low degree of automation of coating appearance detection leads to poor detection optimization efficiency.

[0003] Therefore, in the current electroplating solution coating detection optimization related technologies, the coating appearance inspection relies on experience, which is highly subjective, inefficient and inconsistent, which in turn leads to technical problems such as product quality fluctuations and differences. Summary of the invention

[0004] The present application provides an optimization method for coating appearance inspection of electroplating solutions, and adopts technical means such as component fitness evaluation function and parameter optimization to solve the technical problems of existing electroplating solution coating appearance inspection that relies on experience, is highly subjective, inefficient and inconsistent, and thus leads to fluctuations and differences in product quality. The application realizes accurate, efficient and automated coating appearance inspection, and achieves the technical effect of improving product quality stability and consistency.

[0005] The present application provides a method for optimizing the appearance detection of a coating of an electroplating solution, the method comprising: performing appearance quality detection on a coating sample to obtain a color deviation coefficient, a thickness deviation coefficient and a roughness deviation coefficient; when the color deviation coefficient is greater than or equal to a first deviation coefficient threshold, or / and the thickness deviation coefficient is greater than or equal to a second deviation coefficient threshold, or / and the roughness deviation coefficient is greater than or equal to a third deviation coefficient threshold, obtaining electroplating control parameters of the coating sample, wherein the electroplating control parameters include an initial pH value of the electroplating solution, an initial proportion of the electroplating solution components, an initial temperature of the electroplating solution and an initial current density; constructing a fitness evaluation function according to a deviation index, wherein the fitness evaluation function is a maximum value sorting function; according to the fitness evaluation function, optimizing the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density to obtain an optimization result of the electroplating control parameters; and optimizing the coating appearance according to the optimization result of the electroplating control parameters.

[0006] In a possible implementation, an appearance quality inspection is performed on the coating sample to obtain a color deviation coefficient, and the following processing is performed: image acquisition is performed on the coating sample to obtain a coating digital image; pixel-level color region clustering is performed on the coating digital image to obtain a first region color feature value, a second region color feature value, and so on, up to the Nth region color feature value, where N is an integer and N≥1; variance calculation is performed on the first region color feature value, the second region color feature value, and so on, up to the Nth region color feature value to obtain a color distribution dispersion coefficient; when the color distribution dispersion coefficient is greater than or equal to a color dispersion coefficient threshold, the color deviation coefficient is set to 1; when the color distribution dispersion coefficient is less than the color dispersion coefficient threshold, the proportion of regions in which the deviations of the first region color feature value, the second region color feature value, and so on, up to the Nth region color feature value from the reference color feature value are greater than or equal to the color deviation threshold is counted, and is set as the color deviation coefficient.

[0007] In a possible implementation, pixel-level color region clustering is performed on the coating digital image to obtain a first region color feature value, a second region color feature value, and so on until the Nth region color feature value, and the following processing is performed: from the coating digital image, an RGB feature value of a first comparison region and an RGB feature value of a second comparison region are extracted, wherein the first comparison region and the second comparison region are adjacent regions; the RGB feature value of the first comparison region and the RGB feature value of the second comparison region are compared to obtain an RGB feature value deviation; when the RGB feature value deviation is less than an RGB feature value deviation threshold, the number of pixels in the first region and the number of pixels in the second region are obtained; the sum of the number of pixels in the first region and the number of pixels in the second region is calculated to obtain an image The result of adding the number of pixels in the first area is compared with the result of adding the number of pixels, and the result is set as the first area weight; the result of adding the number of pixels in the second area is compared with the result of adding the number of pixels, and the result is set as the second area weight; according to the first area weight and the second area weight, the weighted mean of the RGB feature value of the first comparison area and the RGB feature value of the second comparison area is calculated to obtain the RGB feature value of the aggregated area, wherein the aggregated area is equal to the union of the first area and the second area; repeat the analysis, and stop when the deviation of the RGB feature values ​​of any two areas is greater than or equal to the RGB feature value deviation threshold, and output the color feature value of the first area, the color feature value of the second area until the color feature value of the Nth area.

[0008] In a possible implementation, the coating sample is subjected to appearance quality inspection to obtain a thickness deviation coefficient, and the following processing is also performed: the coating sample is subjected to a plurality of uniform point detections by an X-ray film thickness detector to obtain a plurality of film thickness detection results; variance calculation is performed on the plurality of film thickness detection results to obtain a film thickness distribution dispersion coefficient; when the film thickness distribution dispersion coefficient is greater than or equal to a film thickness dispersion coefficient threshold, the thickness deviation coefficient is set to 1; when the film thickness distribution dispersion coefficient is less than the film thickness dispersion coefficient threshold, the proportion of the plurality of film thickness detection results whose deviations from the reference film thickness characteristic value are greater than or equal to the film thickness deviation threshold is counted and set as the thickness deviation coefficient.

[0009] In a possible implementation, the appearance quality of the coating sample is inspected to obtain a roughness deviation coefficient, and the following processing is performed: the coating sample is imaged and grayed to obtain a coating grayscale digital image; the coating grayscale digital image is clustered at the pixel level grayscale region to obtain a first region grayscale eigenvalue, a second region grayscale eigenvalue, and so on, up to the Qth region grayscale eigenvalue, where Q is an integer and Q≥1; a roughness grayscale threshold is configured; the regions of the first region grayscale eigenvalue, the second region grayscale eigenvalue, and so on, which are less than or equal to the roughness grayscale threshold, are positionally identified to obtain a grayscale area to be inspected; when the ratio of the number of grayscale areas to be inspected to Q is greater than or equal to the grayscale ratio threshold, the roughness deviation coefficient is set to 1.

[0010] In a possible implementation, the coating sample is subjected to appearance quality inspection to obtain a roughness deviation coefficient, and the following processing is also performed: when the ratio of the number of grayscale areas to be inspected in the grayscale area to be inspected to Q is less than the grayscale ratio threshold, the grayscale area to be inspected is subjected to roughness inspection through a light cutting microscope to obtain a roughness detection area; the ratio of the number of roughness detection areas in the roughness detection area to Q is calculated and set as the roughness deviation coefficient.

[0011] In a possible implementation, a fitness evaluation function is constructed according to the deviation index, and the following processing is also performed: the fitness evaluation function is: ; in, Represents the fitness evaluation value, The deviation coefficient threshold that characterizes the deviation index of the i-th dimension, represents the eigenvalue of the deviation index of the i-th dimension, and M represents the dimension of the deviation index.

[0012] In a possible implementation, according to the fitness evaluation function, the initial pH value of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density are optimized to obtain the optimization result of the electroplating control parameters, and the following processing is performed: based on the Internet of Things, the initial pH value of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density are assigned to obtain several groups of retrieved electroplating control parameters; according to the fitness evaluation function, several fitness evaluation values ​​of the several groups of retrieved electroplating control parameters are obtained; according to the several fitness evaluation values, the several groups of retrieved electroplating control parameters are sorted for the maximum values ​​to obtain the optimization result of the electroplating control parameters.

[0013] It is intended to conduct appearance quality inspection of coating samples through a coating appearance inspection optimization method of an electroplating solution proposed in this application; the color deviation coefficient is greater than or equal to the first deviation coefficient threshold, or / and the thickness deviation coefficient is greater than or equal to the second deviation coefficient threshold, or / and the roughness deviation coefficient is greater than or equal to the third deviation coefficient threshold, and the electroplating control parameters of the coating samples are obtained; according to the deviation index, a fitness evaluation function is constructed, wherein the fitness evaluation function is a maximum value sorting function; according to the fitness evaluation function, the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density are optimized to obtain the optimization results of the electroplating control parameters; the coating appearance is optimized according to the optimization results of the electroplating control parameters. It solves the technical problems of existing electroplating solution coating inspection that relies on experience for coating appearance inspection, is highly subjective, inefficient and has poor consistency, which leads to fluctuations and differences in product quality, and realizes accurate, efficient and automated coating appearance inspection, achieving the technical effect of improving product quality stability and consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic flow chart of a method for optimizing the appearance detection of a coating of an electroplating solution provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for obtaining a color deviation coefficient in a method for optimizing appearance detection of a coating of an electroplating solution provided in an embodiment of the present application; Figure 3A schematic diagram of a process for obtaining a thickness deviation coefficient in a method for optimizing appearance detection of a coating of an electroplating solution provided in an embodiment of the present application; Figure 4 A schematic diagram of a process for obtaining a roughness deviation coefficient in a method for optimizing appearance detection of a coating of an electroplating solution provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0017] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0018] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0019] The present application embodiment provides a method for optimizing the appearance detection of a coating of an electroplating solution, such as Figure 1 As shown, the method includes: Step S100, performing appearance quality inspection on the coating sample to obtain a color deviation coefficient, a thickness deviation coefficient and a roughness deviation coefficient. In the appearance quality inspection of coating samples, the color deviation coefficient, thickness deviation coefficient and roughness deviation coefficient are all important parameters used to measure the quality of coatings. Specifically, the color deviation coefficient is used to quantify the difference between the coating color and the expected standard coating color. Usually, the color deviation is accurately measured by instruments such as a colorimeter or a spectrometer, and the hue, lightness and saturation of the coating sample are compared with the hue, lightness and saturation of the standard sample to determine the color of the coating of automotive parts, decorations, etc. The color deviation coefficient is used to accurately control the color of the coating; the thickness deviation coefficient is used to quantify the difference between the actual thickness of the coating and the expected standard thickness. The thickness of the coating has a direct impact on its performance (for example, corrosion resistance, conductivity, etc.). Usually, a thickness measuring instrument (for example, a coating thickness gauge) is used to measure the coating thickness and compare it with the standard coating thickness; the roughness deviation coefficient is used to quantify the difference between the actual roughness of the coating surface and the expected standard roughness. Roughness is an important factor affecting the appearance quality and performance of the coating. Roughness is usually measured using a surface roughness measuring instrument, such as a profilometer or an optical microscope.

[0020] In one possible implementation, Figure 2As shown, step S100, performing appearance quality inspection on the coating sample to obtain the color deviation coefficient, and also includes step S110, performing image acquisition on the coating sample to obtain a digital image of the coating. Use a digital image acquisition device (such as a camera, scanner or microscope) to shoot or scan the coating surface to obtain digital image data of the coating. Specifically, ensure that the surface of the coating sample is clean and free of debris, and place it on a platform suitable for image acquisition. For some samples that require a specific observation angle or magnification, it may be necessary to adjust the position of the sample or use a special fixture to fix it. According to the characteristics of the coating and the detection requirements, select a suitable image acquisition device. For example, for microscopic observation of the coating sample, it may be necessary to use a high-resolution microscope or scanning electron microscope. For macroscopic image data acquisition of the coating sample, use an ordinary camera or scanner to shoot or scan the coating sample, and convert the surface morphology, color, texture and other characteristics of the coating into digital image data, that is, obtain a digital image of the coating. The method further includes step S120, in which the color region clustering of the coating digital image is performed at the pixel level to obtain the color feature value of the first region, the color feature value of the second region, and the color feature value of the Nth region, wherein N is an integer and N≥1. The pixels in the coating digital image are clustered and grouped according to the color features, so as to identify a plurality of different regions with similar color features. Each clustered region represents a part of the image with similar colors, which is helpful for analyzing and understanding the distribution and characteristics of different color regions in the image. Specifically, color features are extracted from each pixel of the image, and the extracted color features are clustered, for example, using K-means, mean clustering, hierarchical clustering, etc. For each color region obtained by clustering, its color feature value is calculated, wherein the color feature value of the first region, the color feature value of the second region, and the color feature value of the Nth region refer to the specific values ​​of the color features represented by each clustering region, for example, the median or mean of the color feature value is calculated. The color features of different regions in the coating digital image are obtained, so as to identify abnormal color regions in the coating, analyze the uniformity of color distribution, compare color differences between different samples, etc. The method further includes step S130 of performing variance calculation on the color feature values ​​of the first region, the color feature values ​​of the second region, and the color feature values ​​of the Nth region to obtain a color distribution dispersion coefficient.For each color region (from the first region to the Nth region), the average value of its color feature value is calculated. For each sample point (pixel) in each color region, the deviation of its color feature value from the regional average value is calculated to obtain the sum of squares of the deviations of each color region, which is divided by the number of sample points (i.e., the number of pixels) in the region to obtain the variance of each color region, and the color distribution dispersion coefficient is determined. The calculation formula of the dispersion coefficient is usually to divide the standard deviation (square root of the variance) of the color region by the average value, which describes the uniformity of the distribution of color feature values ​​in different regions. For example, a lower dispersion coefficient indicates that the color distribution in the region is relatively uniform, and a higher dispersion coefficient indicates that the color distribution in the region is uneven. It also includes step S140, when the color distribution dispersion coefficient is greater than or equal to the color dispersion coefficient threshold, the color deviation coefficient is set to 1. Among them, the color dispersion coefficient threshold is a preset value used to determine whether the degree of dispersion of the color distribution exceeds the range. When the color distribution dispersion coefficient is greater than or equal to the color dispersion coefficient threshold, it indicates that the color distribution in the region is uneven, and the color deviation coefficient is set to 1. The method further includes step S150, in which, when the color distribution dispersion coefficient is less than the color dispersion coefficient threshold, the color feature value of the first region, the color feature value of the second region, and the area ratio where the deviation between the color feature value of the Nth region and the reference color feature value is greater than or equal to the color deviation threshold are counted, and the area ratio is set as the color deviation coefficient. When the color distribution dispersion coefficient is less than the color dispersion coefficient threshold, it indicates that the color distribution is relatively uniform and there is no significant discrete phenomenon, but there may be a large deviation between the color feature value of some regions and the reference color feature value. Therefore, it is necessary to further analyze the deviation between the color feature value of each region and the reference color feature value, and count the area ratio where the deviation is greater than or equal to the color deviation threshold, and set it as the color deviation coefficient to quantify the degree of color deviation.

[0021] In a possible implementation, step S120 further includes step S121, extracting the RGB feature value of the first comparison area and the RGB feature value of the second comparison area from the coating digital image, wherein the first comparison area and the second comparison area are adjacent areas. Feature extraction in the RGB color space is performed on two adjacent areas in the image (i.e., the first comparison area and the second comparison area), wherein RGB (red, green, and blue) is a color space that represents color by combining the three primary colors of red, green, and blue with different intensities, and the first comparison area and the second comparison area are two random adjacent areas. For each comparison area, its RGB feature value is extracted, and extracting the RGB feature value of the adjacent area is helpful for analyzing the color change and distribution of the coating sample in a local range. It also includes step S122, comparing the RGB feature value of the first comparison area with the RGB feature value of the second comparison area to obtain the RGB feature value deviation. The feature values ​​of the two adjacent areas in the RGB color space are compared, the difference or deviation between them is calculated, and the difference in color between the two areas is quantified. It also includes step S123, when the RGB feature value deviation is less than the RGB feature value deviation threshold, the number of pixels in the first area and the number of pixels in the second area are obtained. When the RGB feature value deviation is less than the RGB feature value deviation threshold, it means that the difference in color between the two comparison areas (the first comparison area and the second comparison area) is within an acceptable range, that is, their colors are similar, wherein the RGB feature value deviation threshold is a preset value used to determine whether the feature value deviation of the two comparison areas in the RGB color space exceeds the allowable range. Obtaining the number of pixels in the first area and the number of pixels in the second area refers to clarifying the precise boundaries of the first comparison area and the second comparison area in the image. For each comparison area, the number of pixels contained in each area is calculated. Specifically, the pixels in the image are traversed, and each pixel in the area is counted into the total number of pixels in the area to obtain the number of pixels in the first area and the number of pixels in the second area. It also includes step S124, calculating the sum of the number of pixels in the first area and the number of pixels in the second area, obtaining the sum of the number of pixels, comparing the number of pixels in the first area with the sum of the number of pixels, and setting it as the first area weight, and comparing the number of pixels in the second area with the sum of the number of pixels, and setting it as the second area weight. The method further includes step S125, performing weighted mean calculation on the first comparison area RGB feature value and the second comparison area RGB feature value according to the first area weight and the second area weight to obtain the aggregated area RGB feature value, wherein the aggregated area is equal to the union of the first area and the second area.Using the first area weight and the second area weight, the RGB feature values ​​of the first comparison area and the second comparison area are weighted averaged to obtain the aggregated area RGB feature value, the aggregated area RGB feature value = (first area weight × first area RGB feature value + second area weight * second area RGB feature value) / (first area weight + second area weight), wherein the aggregated area is equal to the union of the first area and the second area, that is, the area containing all pixels in the two areas. It also includes step S126, repeating the analysis, stopping when the deviation of the RGB feature values ​​of any two areas is greater than or equal to the RGB feature value deviation threshold, outputting the first area color feature value, the second area color feature value until the Nth area color feature value. If the deviation is less than the threshold, the next area is set as the current area, and the analysis is repeated, the RGB feature value of the area is continued to be extracted and compared with the next adjacent area. If in a certain comparison, it is found that the deviation of the RGB feature values ​​of any two adjacent areas is greater than or equal to the threshold, the analysis is stopped and the color feature values ​​of the 1st to Nth areas are output.

[0022] In one possible implementation, Figure 3As shown, step S100 performs appearance quality inspection on the coating sample to obtain the thickness deviation coefficient, and further includes step S160, performing detection on the coating sample at several uniform points through an X-ray film thickness detector to obtain several film thickness detection results. The X-ray film thickness detector is used to perform non-contact film thickness measurement on the coating sample at several uniform points to obtain several film thickness detection results. Specifically, several uniformly distributed points on the coating sample are selected, and these points are detected one by one using the X-ray film thickness detector. According to the intensity change of the X-ray, the instrument can calculate the film thickness of the coating at each point, and obtain several film thickness detection results, each result corresponding to the film thickness of a detection point. It also includes step S170, performing variance calculation on the several film thickness detection results to obtain the film thickness distribution dispersion coefficient. The variance of several film thickness test results is calculated, wherein the calculation formula is: variance = [(film thickness test result - average film thickness)²] / number of data points. The larger the variance, the greater the difference between the film thickness test results, that is, the more uneven the film thickness distribution; conversely, the smaller the variance, the more uniform the film thickness distribution. The film thickness distribution dispersion coefficient is further calculated, which is the ratio of the variance to the average value, and is used to compare the degree of dispersion of the film thickness distribution. It also includes step S180, when the film thickness distribution dispersion coefficient is greater than or equal to the film thickness dispersion coefficient threshold, the thickness deviation coefficient is set to 1. When the film thickness distribution dispersion coefficient is greater than or equal to the film thickness dispersion coefficient threshold, it indicates that there is a significant difference in the distribution of the film thickness, which may exceed the allowable range and affect the quality and performance of the product. The thickness deviation coefficient is set to 1, wherein the film thickness dispersion coefficient threshold is a preset value used to determine whether the dispersion degree of the film thickness distribution of the coating sample exceeds the range. The method further includes step S190, wherein when the film thickness distribution dispersion coefficient is less than the film thickness dispersion coefficient threshold, the proportion of the deviations of the plurality of film thickness detection results from the reference film thickness characteristic value greater than or equal to the film thickness deviation threshold is counted and set as the thickness deviation coefficient. When the film thickness distribution dispersion coefficient is less than the film thickness dispersion coefficient threshold, it means that the film thickness distribution of the coating sample is relatively uniform and no significant discrete phenomenon occurs. However, even if the overall distribution is uniform, there may be a large deviation between individual film thickness detection results and the reference film thickness characteristic value, wherein the reference film thickness characteristic value is the expected standard film thickness value. Specifically, each film thickness detection result is compared with the reference film thickness characteristic value, and the deviation therebetween is calculated. If the deviation of a certain film thickness detection result from the reference value is greater than or equal to the preset film thickness deviation threshold, it means that the detection result has a significant film thickness deviation. The number of film thickness detection results with a deviation greater than or equal to the film thickness deviation threshold is counted among all film thickness detection results, and the proportion of this number to the total number of all film thickness detection results is calculated as the thickness deviation coefficient.

[0023] In one possible implementation, Figure 4As shown, step S100 performs appearance quality inspection on the coating sample to obtain the roughness deviation coefficient, and further includes step S1100, collecting images of the coating sample and graying it to obtain a grayscale digital image of the coating. The coating sample is photographed or scanned by an image acquisition device (such as a camera, a scanner, etc.) to obtain its original color image, and the collected color image is grayed to obtain a grayscale digital image of the coating, wherein graying is an image processing technology that converts a color image into a grayscale image. In the graying process, each pixel of the image is converted from the original RGB (red, green, and blue) three color channel values ​​to a grayscale value, which represents the brightness information of the pixel. Through graying, the color information in the image can be removed, and only the brightness information is retained. It also includes step S1110, pixel-level grayscale region clustering of the coating grayscale digital image is performed to obtain the grayscale eigenvalue of the first region, the grayscale eigenvalue of the second region, and the grayscale eigenvalue of the Qth region, wherein Q is an integer, and Q≥1. Among them, pixel-level grayscale region clustering is to cluster pixels with similar grayscale values ​​together to form different regions. In grayscale images, the value of pixels represents brightness or grayscale level. Pixel clustering can identify pixels with similar brightness characteristics and classify them into the same group, obtain the first region grayscale feature value, the second region grayscale feature value until the Qth region grayscale feature value. Specifically, by performing pixel-level grayscale region clustering on the grayscale digital image of the coating, several different grayscale regions are obtained, each region represents a different grayscale feature of the coating surface, and the grayscale feature value of each region is calculated, for example, the average value or median of the grayscale feature values ​​of all pixels in the region is calculated, representing the overall grayscale level of the region, which is used to further analyze the grayscale distribution characteristics of the coating surface, such as detecting whether there is an abnormal area, evaluating the uniformity of the coating, etc. It also includes step S1130, configuring a rough grayscale threshold. Specifically, in image processing, a specific brightness value is set as a threshold to divide the grayscale level of the image into two categories above and below the threshold, that is, into two categories of grayscale images with different roughness. The method further includes step S1140, in which the grayscale characteristic value of the first region, the grayscale characteristic value of the second region, and the grayscale characteristic value of the Qth region are less than or equal to the rough grayscale threshold, and the grayscale inspection region is obtained. The grayscale characteristic value of the first to Q regions is less than or equal to the rough grayscale threshold, and the region to be inspected is identified to generate the grayscale inspection region. The method further includes step S1150, in which when the ratio of the number of grayscale inspection regions to Q of the grayscale inspection region is greater than or equal to the grayscale ratio threshold, the roughness deviation coefficient is set to 1.When the ratio of the number of grayscale areas to be inspected to Q is greater than or equal to the grayscale ratio threshold, the roughness deviation coefficient is set to 1. Specifically, the ratio of the number of grayscale areas to be inspected to Q is calculated. This ratio represents the proportion of abnormal areas in the entire image. If this ratio is greater than or equal to the preset grayscale ratio threshold, it means that there are more problem areas in the image and further processing is required, where Q is the total number of areas obtained after pixel-level grayscale area clustering, that is, the total number of different grayscale areas segmented in the image; the grayscale ratio threshold is a parameter for setting the proportion of problem areas in an image. If the ratio exceeds this threshold, it means that there is an abnormality in the grayscale digital image of the coating.

[0024] In a possible implementation, step S1150 also includes step S1151, when the ratio of the number of grayscale areas to be inspected in the grayscale area to be inspected to Q is less than the grayscale ratio threshold, the grayscale area to be inspected is subjected to roughness detection by a light sectioning microscope to obtain a roughness detection area. When the ratio of the number of grayscale areas to be inspected in the grayscale area to be inspected to Q is less than the grayscale ratio threshold, it indicates that there are relatively few abnormal areas in the grayscale digital image of the coating, and the grayscale area to be inspected is subjected to roughness detection by a light sectioning microscope to obtain a roughness detection area, wherein the light sectioning microscope measures the surface roughness using the principle of light sectioning method, specifically, the grayscale area to be inspected is subjected to roughness detection by a light sectioning microscope, the surface roughness of these areas is accurately measured, and the roughness detection area is determined. It also includes step S1152, calculating the ratio of the number of roughness detection areas in the roughness detection area to Q, and setting it as the roughness deviation coefficient. Calculate the ratio of the number of roughness detection areas to Q, that is, calculate the ratio of the number of roughness detection areas to the total number of areas, which reflects how many proportions of areas have roughness deviations in all areas that have been clustered by grayscale areas. The lower the ratio, the fewer areas of the coating sample with roughness deviations, and the better the sample quality. This ratio is set as the roughness deviation coefficient.

[0025] Step S200, when the color deviation coefficient is greater than or equal to the first deviation coefficient threshold, or / and the thickness deviation coefficient is greater than or equal to the second deviation coefficient threshold, or / and the roughness deviation coefficient is greater than or equal to the third deviation coefficient threshold, the electroplating control parameters of the coating sample are obtained, wherein the electroplating control parameters include the initial pH value of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution, and the initial current density. When the color deviation coefficient is greater than or equal to the first deviation coefficient threshold, it means that the difference between the color of the coating and the expected standard color exceeds the allowable range; when the thickness deviation coefficient is greater than or equal to the second deviation coefficient threshold, it means that the thickness distribution of the coating is uneven or exceeds the expected standard thickness range; when the roughness deviation coefficient is greater than or equal to the third deviation coefficient threshold, it means that the roughness of the coating surface is too high and exceeds the expected standard roughness; wherein the first deviation coefficient threshold is the value of the preset coating color deviation coefficient, the second deviation coefficient threshold is the value of the preset coating thickness deviation coefficient, and the first deviation coefficient threshold is the value of the preset coating roughness deviation coefficient. If at least one of the color deviation coefficient, thickness deviation coefficient and roughness deviation coefficient is greater than or equal to the corresponding deviation coefficient threshold, it means that there is an abnormality in the appearance quality of the coating, and it is necessary to obtain the relevant parameter values ​​of the coating sample during electroplating, that is, the electroplating control parameters, including the initial pH value of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density. Specifically, the pH value affects the formation speed, uniformity and corrosion resistance of the coating; different concentration combinations of metal ions and additives will produce different plating effects; temperature affects the electroplating reaction rate and ion diffusion rate, thereby affecting the coating quality; current density directly affects the growth rate and thickness distribution of the coating. Adjusting these electroplating control parameters can optimize the electroplating process and improve the coating quality.

[0026] Step S300, constructing a fitness evaluation function according to the deviation index, wherein the fitness evaluation function is a maximum value sorting function. According to deviation indexes such as color deviation coefficient, thickness deviation coefficient and roughness deviation coefficient, a fitness evaluation function is constructed to evaluate the fitness of the control parameters of the electroplating process, wherein the fitness evaluation function is a maximum value sorting function, and the goal is to find the maximum fitness value (the optimal combination of electroplating control parameters). The maximum value sorting function is a special type of fitness evaluation function. When optimizing the electroplating process, a set of electroplating control parameters is found based on the fitness evaluation function, so that the deviation indexes such as color deviation, thickness deviation and roughness deviation are minimized, thereby achieving the optimal coating quality.

[0027] In a possible implementation, step S300 further includes step S310, and the fitness evaluation function is: ; in, Represents the fitness evaluation value, The deviation coefficient threshold that characterizes the deviation index of the i-th dimension, represents the eigenvalue of the deviation index of the i-th dimension, and M represents the dimension of the deviation index.

[0028] Step S400, according to the fitness evaluation function, the initial value of the pH of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density are optimized to obtain the optimization result of the plating control parameters. According to the fitness evaluation function, the initial value of the pH of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density are optimized, that is, the fitness evaluation function is used to systematically search and determine the best combination of plating control parameters. Specifically, the fitness evaluation function is used to calculate the fitness value of the parameter combination according to deviation indicators such as color deviation coefficient, thickness deviation coefficient and roughness deviation coefficient, and the fitness is repeatedly evaluated and calculated to obtain one or more groups of plating control parameters (including the initial value of the pH of the plating solution, the initial proportion of the plating solution components, the initial temperature of the plating solution and the initial current density), which show the best performance under the fitness evaluation function, that is, it can minimize deviation indicators such as color deviation, thickness deviation and roughness deviation, and achieve high-quality plating effect.

[0029] In a possible implementation, step S400 further includes step S410, based on the Internet of Things, assigning the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density, and obtaining several groups of retrieval electroplating control parameters. Based on the Internet of Things technology, the pH value of the electroplating solution, the proportion of the electroplating solution components, the temperature of the electroplating solution and the current density in the electroplating production process are collected in real time to obtain several groups of retrieval electroplating control parameters. It also includes step S420, according to the fitness evaluation function, obtaining several fitness evaluation values ​​of the several groups of retrieval electroplating control parameters. Substitute the several groups of electroplating control parameters assigned by the Internet of Things technology into the fitness evaluation function for calculation, so as to obtain the fitness value corresponding to each group of parameters. It also includes step S430, according to the several fitness evaluation values, the several groups of retrieval electroplating control parameters are sorted for the maximum value, and the optimization result of the electroplating control parameters is obtained. Among all the retrieved electroplating control parameter combinations, by comparing their fitness evaluation values, the set of parameters with the maximum fitness value is selected as the optimization result, so as to optimize the electroplating process and improve the appearance inspection efficiency and coating quality of the electroplated coating.

[0030] Step S500, optimizing the appearance of the coating according to the electroplating control parameter optimization result. The optimal electroplating control parameter combination (including the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density) selected by the fitness evaluation function is used to optimize and adjust the actual electroplating process to improve the quality of the coating.

[0031] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for optimizing the appearance detection of a coating of an electroplating solution, characterized in that: include: Conduct appearance quality inspection on coating samples to obtain color deviation coefficient, thickness deviation coefficient and roughness deviation coefficient; When the color deviation coefficient is greater than or equal to the first deviation coefficient threshold, or / and the thickness deviation coefficient is greater than or equal to the second deviation coefficient threshold, or / and the roughness deviation coefficient is greater than or equal to the third deviation coefficient threshold, the electroplating control parameters of the coating sample are obtained, wherein the electroplating control parameters include an initial pH value of the electroplating solution, an initial proportion of the electroplating solution components, an initial temperature of the electroplating solution, and an initial current density; According to the deviation index, a fitness evaluation function is constructed, wherein the fitness evaluation function is a maximum value sorting function; According to the fitness evaluation function, optimizing the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density to obtain an optimization result of the electroplating control parameters; The coating appearance is optimized according to the electroplating control parameter optimization result.

2. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 1, characterized in that: Conduct appearance quality inspection on coating samples to obtain color deviation coefficients, including: Capturing images of the coating sample to obtain a digital image of the coating; Performing pixel-level color region clustering on the coating digital image to obtain a first region color feature value, a second region color feature value, and finally an Nth region color feature value, wherein N is an integer, and N≥1; Performing variance calculation on the color feature values ​​of the first region, the color feature values ​​of the second region, and up to the color feature values ​​of the Nth region to obtain a color distribution dispersion coefficient; When the color distribution dispersion coefficient is greater than or equal to the color dispersion coefficient threshold, the color deviation coefficient is set to 1; When the color distribution dispersion coefficient is less than the color dispersion coefficient threshold, the color feature values ​​of the first area and the second area are counted until the deviation between the color feature value of the Nth area and the reference color feature value is greater than or equal to the color deviation threshold, which is set as the color deviation coefficient.

3. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 2, characterized in that: Performing pixel-level color region clustering on the coating digital image to obtain a first region color feature value, a second region color feature value, and finally an Nth region color feature value, comprises: Extracting the RGB feature value of the first comparison area and the RGB feature value of the second comparison area from the coating digital image, wherein the first comparison area and the second comparison area are adjacent areas; Comparing the RGB feature value of the first comparison area with the RGB feature value of the second comparison area to obtain an RGB feature value deviation; When the RGB feature value deviation is less than the RGB feature value deviation threshold, obtaining the number of pixels in the first area and the number of pixels in the second area; Calculate the sum of the number of pixels in the first area and the number of pixels in the second area to obtain a pixel sum result, compare the number of pixels in the first area with the pixel sum result, and set it as a first area weight; compare the number of pixels in the second area with the pixel sum result, and set it as a second area weight; According to the first region weight and the second region weight, weighted mean calculation is performed on the RGB feature value of the first comparison region and the RGB feature value of the second comparison region to obtain the RGB feature value of the aggregated region, wherein the aggregated region is equal to the union of the first region and the second region; Repeat the analysis, and stop when the deviation of the RGB feature values ​​of any two regions is greater than or equal to the RGB feature value deviation threshold, and output the first region color feature value, the second region color feature value, and so on until the Nth region color feature value.

4. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 1, characterized in that: Conduct appearance quality inspection on the coating samples to obtain the thickness deviation coefficient, including: Performing a plurality of uniform point detections on the coating sample by an X-ray film thickness detector to obtain a plurality of film thickness detection results; Performing variance calculation on the plurality of film thickness detection results to obtain a film thickness distribution dispersion coefficient; When the film thickness distribution dispersion coefficient is greater than or equal to the film thickness dispersion coefficient threshold, the thickness deviation coefficient is set to 1; When the film thickness distribution dispersion coefficient is less than the film thickness dispersion coefficient threshold, the proportion of the deviations between the plurality of film thickness detection results and the reference film thickness characteristic values ​​that are greater than or equal to the film thickness deviation threshold is counted and set as the thickness deviation coefficient.

5. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 1, characterized in that: Perform appearance quality inspection on the coating samples to obtain the roughness deviation coefficient, including: Capturing and graying the coating sample to obtain a grayscale digital image of the coating; Performing pixel-level grayscale regional clustering on the coating grayscale digital image to obtain a first region grayscale eigenvalue, a second region grayscale eigenvalue, and finally a Qth region grayscale eigenvalue, wherein Q is an integer, and Q≥1; Configure the rough grayscale threshold; Mark the position of the region where the grayscale characteristic value of the first region, the grayscale characteristic value of the second region, and the grayscale characteristic value of the Qth region are less than or equal to the rough grayscale threshold, to obtain a grayscale inspection region; When the ratio of the number of grayscale areas to be inspected to Q of the grayscale areas to be inspected is greater than or equal to the grayscale ratio threshold, the roughness deviation coefficient is set to 1.

6. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 5, characterized in that: Also includes: When the ratio of the number of grayscale areas to be inspected to Q of the grayscale areas to be inspected is less than the grayscale ratio threshold, performing roughness detection on the grayscale areas to be inspected by using a light sectioning microscope to obtain a roughness detection area; The ratio of the number of roughness detection areas in the roughness detection area to Q is calculated and set as the roughness deviation coefficient.

7. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 1, characterized in that: According to the deviation index, a fitness evaluation function is constructed, including: The fitness evaluation function is: ； in, Represents the fitness evaluation value, The deviation coefficient threshold that characterizes the deviation index of the i-th dimension, represents the eigenvalue of the deviation index of the i-th dimension, and M represents the dimension of the deviation index.

8. The method for optimizing the appearance detection of a coating of an electroplating solution according to claim 1, characterized in that: According to the fitness evaluation function, the initial pH value of the electroplating solution, the initial proportion of the electroplating solution components, the initial temperature of the electroplating solution and the initial current density are optimized to obtain the optimization result of the electroplating control parameters, including: Based on the Internet of Things, assigning values ​​to the initial pH value of the electroplating solution, the initial proportion of the components of the electroplating solution, the initial temperature of the electroplating solution and the initial current density, to obtain several groups of retrieved electroplating control parameters; According to the fitness evaluation function, obtaining the plurality of fitness evaluation values ​​of the plurality of groups of retrieved electroplating control parameters; The plurality of groups of retrieved electroplating control parameters are sorted for maximum values ​​according to the plurality of fitness evaluation values ​​to obtain the electroplating control parameter optimization result.

Images