A method and system for monitoring the operating status of a metal workpiece pretreatment production line
By dynamically adjusting the spray image limiting parameters using local dark channel values and time window sliding division techniques, the problem of insufficient adaptability of CLAHE limiting parameters under dynamic water mist was solved, achieving efficient cleaning and stable product quality in the metal workpiece pretreatment production line.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG DEQING FUYUAN INTELLIGENT EQUIP TECH CO LTD
- Filing Date
- 2026-02-12
- Publication Date
- 2026-05-26
Smart Images
Figure CN121708015B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image data processing, and in particular to a method and system for monitoring the operating status of a metal workpiece pre-processing production line. Background Technology
[0002] In the pretreatment production line for metal workpieces, spray cleaning is a key step, mainly used to remove contaminants such as grease, dirt, rust and oxides attached to the surface of the workpiece to ensure surface cleanliness and provide a foundation for subsequent processes such as coating, powder coating or enamel treatment.
[0003] In existing technologies, spray pressure is typically controlled manually or by fixing parameters, but this lacks real-time monitoring, resulting in low cleaning efficiency, resource waste, and unstable product quality. Currently, image processing technology has been introduced into production line monitoring. In the monitoring of the operating status of existing metal workpiece pretreatment production lines, when using image processing technology to analyze the characteristics of adhering substances on the workpiece surface, the main reliance is on the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm for image enhancement to highlight contaminant areas.
[0004] However, in actual production line spraying environments, due to contrast attenuation and local noise amplification caused by dynamic water mist, CLAHE's fixed limiting parameters cannot adapt to changes in water mist concentration. For example, excessively high limiting in dense fog areas leads to excessive noise enhancement and artifacts, while excessively low limiting in thin fog areas results in loss of detail. Consequently, it is impossible to adjust the spray pressure accurately in real time, affecting the cleaning effect and production efficiency. Summary of the Invention
[0005] To address the issue that CLAHE's fixed limiting parameters cannot adapt to changes in water mist concentration due to contrast attenuation and local noise amplification caused by dynamic water mist, this application provides a method and system for monitoring the operating status of a metal workpiece pretreatment production line.
[0006] In a first aspect, this application provides a method for monitoring the operating status of a metal workpiece pretreatment production line, employing the following technical solution, including:
[0007] The process involves acquiring a grayscale spray image of the metal workpiece surface, calculating the local grayscale difference of each pixel within a preset neighborhood, using the minimum grayscale value in the neighborhood as the local dark channel value, dividing the spray image into multiple blocks by a grid, and calculating the suppression degree of each block based on the local dark channel value and local grayscale difference of each pixel, sliding the real-time spray image across preset time windows to obtain multiple window time sequences (each window containing multiple frames of spray images), and dividing it into real-time blocks, calculating the average suppression degree of blocks at the same position in consecutive frames within any window time sequence, setting an initial limiting parameter and a reference suppression degree, and correcting the initial limiting parameter based on the difference between the reference suppression degree and the average suppression degree to obtain the adjusted limiting parameter for each block, processing each block based on the adjusted limiting parameter to obtain the deposit density, setting a baseline spray pressure and a reference deposit density value, and correcting the baseline spray pressure based on the difference between the deposit density and the reference deposit density value to obtain the adjusted spray pressure used to adjust the production line operating status.
[0008] Optionally, the method for calculating the local grayscale difference is as follows: calculate the mean of the sum of the squared differences between the grayscale values of each pixel and all pixels in the neighborhood, and use this as the grayscale difference. After taking the square root of the grayscale difference, standard normalize it to obtain the local grayscale difference.
[0009] Optionally, the method of dividing the spray image into multiple blocks by grid is as follows: if the spray image is less than the size of a block, it is filled to the size of a block.
[0010] Optionally, the method for calculating the suppression degree of each block is as follows: calculate the ratio of the local dark channel value of any pixel in each block to the maximum value of the local dark channel values of all pixels as the first ratio; set hyperparameters; use an exponential function to perform negative correlation mapping on the product of the local gray level difference of any pixel and the hyperparameters as the weight factor of the first ratio; calculate the mean of the sum of the products of the first ratio and the weight factor of all pixels in each block to obtain the suppression degree of each block.
[0011] Optionally, the method for dividing the window time sequence is as follows: for any given moment, the spray image acquired at any given moment is taken as the last spray image data in the window time sequence of any given moment, and spray image data of a preset time window length are acquired sequentially forward.
[0012] Optionally, the method for calculating the adjustment limiting parameter of each block is as follows: calculate the hyperbolic tangent function of the difference between the reference suppression degree and the average suppression degree of each block in the window timing, take half of it and calculate the sum with 1, which is taken as the first sum. The product of the first sum and the initial limiting parameter is taken as the adjustment limiting parameter of each block.
[0013] Optionally, the method for calculating the attachment density is as follows: based on adjusting the amplitude limiting parameter, the image of each block is enhanced using a histogram equalization algorithm to obtain the enhanced block image of each block; the enhanced block image is segmented using the Otsu threshold segmentation algorithm to obtain the number of attachment pixels in each enhanced block image; the ratio of the total number of attachment pixels in all enhanced block images to the total number of pixels in all enhanced block images is taken as the attachment density.
[0014] Optionally, the method for calculating the adjusted spray pressure is as follows: calculate the hyperbolic tangent function of the difference between the density of the attached material and the reference value of the attached material density, take half of it and calculate the sum with 1, which is taken as the second sum. The product of the second sum and the reference spray pressure is taken as the adjusted spray pressure.
[0015] Optionally, the suppression level and spray pressure are updated every time window.
[0016] Secondly, this application provides a monitoring system for the operating status of a metal workpiece pretreatment production line, which adopts the following technical solution:
[0017] A metal workpiece pretreatment production line operation status monitoring system includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the metal workpiece pretreatment production line operation status monitoring method described in any one of the claims is implemented.
[0018] The beneficial effect is that the above-mentioned method for monitoring the operating status of a metal workpiece pretreatment production line generates a computer program and stores it in a memory so that it can be loaded and executed by a processor. Thus, a system can be made based on the memory and processor, which is convenient to use.
[0019] This application has the following technical effects:
[0020] Differentiated processing is implemented based on water mist concentration: In dense fog areas, the limiting parameter is appropriately reduced to avoid over-amplification caused by an excessively high limiting threshold, which would lead to noise being over-amplified and the system misjudging water mist particles and obvious blocky artifacts as stubborn deposits, resulting in unnecessary resource waste. In light fog areas, parameters are adjusted according to the degree of contamination. In areas with higher contamination, the limiting parameter is increased to enhance image contrast and prevent fine contaminants from being masked by an excessively small limiting, thus affecting cleaning effectiveness and production efficiency. In areas with lower contamination, the limiting parameter is reduced, adopting the same strategy as in dense fog areas. Through this refined control strategy, the interference of water mist concentration on the dirt identification results can be minimized. Attached Figure Description
[0021] Figure 1 This is a flowchart illustrating a method for monitoring the operating status of a metal workpiece pretreatment production line according to an embodiment of this application.
[0022] Figure 2 This is a structural block diagram of a metal workpiece pretreatment production line operation status monitoring system according to an embodiment of this application. Detailed Implementation
[0023] This application discloses a method for monitoring the operating status of a metal workpiece pretreatment production line, referring to... Figure 1 The process includes steps S1-S4, as follows:
[0024] S1: Obtain a spray image of the metal workpiece surface after grayscale processing, and calculate the local grayscale difference of each pixel in the spray image within a preset neighborhood.
[0025] It should be noted that metal workpieces inevitably acquire grease, rust, dust, or other impurities during production, storage, transportation, and processing. If these contaminants are not removed, they will affect the uniformity and activity of the workpiece surface, leading to poor adhesion and easy peeling of subsequent coatings (such as powder coating), and even causing product defects. Spray cleaning of the workpiece creates a clean and activated base, which is beneficial for the uniform adhesion and curing of the coating. If cleaning is incomplete, residual oil on the metal workpiece surface can cause blistering, peeling, or uneven coloring of the coating, affecting the product's corrosion resistance and aesthetics.
[0026] In one embodiment, a light source and an industrial camera are installed at a suitable position in the rear section of the spray zone (e.g., directly above a metal workpiece production line) to acquire color images of the workpiece surface in real time.
[0027] The preset sampling frequency calculation method can be:
[0028] ;
[0029] In the formula, Indicates the camera's acquisition frequency; Indicates the production line speed (mm / s); A preset empirical value is set to the desired number of frames to be captured for each workpiece. =3; This is the minimum length (in millimeters) of the workpiece.
[0030] For example, pipeline speed mm / s; minimum workpiece length Millimeters; expected number of frames to be acquired for each workpiece Then the camera's acquisition frequency .
[0031] In this step, an industrial camera is used to capture physical environmental data including the workpiece surface, water mist, and lighting. The RGB (Red-Green-Blue) image matrix (i.e., surface color image).
[0032] The basic structure of an RGB image matrix is: the size of an RGB image is usually expressed in terms of height ( (Number of rows in pixels) × Width ( It is represented by (number of pixel columns) × number of channels (3: corresponding to the three color channels R, G, and B respectively).
[0033] This existing technology is based on optical imaging and the CCD / CMOS (Charge-Coupled Device / Complementary Metal-Oxide-Semiconductor Image Sensor) sensor principle, converting reflected photon streams into digital signals.
[0034] The acquired color image of the workpiece surface is converted to grayscale to obtain a grayscale image, which is then used as the spray image for subsequent processing.
[0035] This completes the acquisition of spray images for the pre-treatment of workpieces on the production line.
[0036] Specifically, for any spray image, the minimum gray value within the eight-neighborhood range of any pixel in the spray image is obtained and used as the local dark channel value of any pixel.
[0037] The mean of the sum of the squared differences between the gray values of each pixel and all pixels in its neighborhood is calculated as the gray-level difference. The local gray-level difference is obtained by taking the square root of the gray-level difference and then standardizing it.
[0038] The formula for calculating the local grayscale difference of any of the above pixels can be:
[0039] ;
[0040] Represents any pixel point Local grayscale differences; Represents any pixel point The total number of neighborhood ranges; Indicates the pixel ordinal number; Represents any pixel point The ordinal number of the pixels within the neighborhood; Represents any pixel point grayscale value; Represents any pixel point Within the neighborhood, the first The grayscale value of each pixel; This represents the standard normalization function.
[0041] in, It represents the magnitude of the difference between any pixel and its neighboring pixels; for any pixel, the greater the difference between the gray value of any pixel and the gray value of its neighboring pixels, the greater the local difference of that pixel.
[0042] S2: Take the minimum gray value in the neighborhood of each pixel as the local dark channel value of each pixel; divide the spray image into multiple blocks according to the grid, and calculate the suppression degree of each block based on the local dark channel value and local gray value difference of each pixel in each block.
[0043] It should be noted that in the pretreatment production line for metal workpieces, the spray cleaning effect is highly dependent on the actual degree of contamination on the surface of the workpiece: when the contamination is heavy, the pressure needs to be increased to thoroughly clean and remove stubborn deposits, while when the contamination is light, the pressure needs to be reduced to avoid energy waste and damage to the workpiece surface.
[0044] During the cleaning process of metal workpieces, the spraying operation removes adhering substances, resulting in the generation of a large amount of water mist on the spray cleaning production line. For example, the water mist is densest directly below the nozzle and sparser at the edges. This water mist causes strong scattered light when capturing spray images, leading to a sharp decrease in the overall image contrast. (Contrast is a core indicator measuring the difference in brightness between bright and dark areas of an image, directly determining the distinguishability between contaminants and the substrate on the workpiece surface. Essentially, contrast is the difference in grayscale values between the brightest and darkest areas of an image.) The greater the difference, the more distinct the image layers and the clearer the details; the smaller the difference, the more hazy the image and the more blurred the features. Local high-frequency noise (water droplet reflection) and artifacts (fog clusters mistaken for dirt patches) appear. If these are not analyzed, during image enhancement: noise in dense fog areas is excessively amplified, causing the system to misjudge water droplets as stubborn deposits, leading to an incorrect increase in spray pressure and wasted resources; contrast enhancement in thin fog areas is insufficient, and fine grease residues are "submerged" in the background and cannot be detected, resulting in insufficient pressure adjustment. This allows workpieces to carry contaminants into subsequent processes, leading to poor layer adhesion. Therefore, this application analyzes the spray images during the spraying process to obtain the degree of suppression.
[0045] In one embodiment, any spray image is divided into grids of the same size. The blocks, such as If the spray image is smaller than a block size, it is filled with mirror image to the size of a block; the degree of suppression for each block is calculated.
[0046] The ratio of the local dark channel value of any pixel in each block to the maximum local dark channel value of all pixels is used as the first ratio. Hyperparameters are set, and an exponential function is used to negatively correlate the product of the local gray level difference of any pixel with the hyperparameter, which is used as the weight factor of the first ratio. The average of the sum of the products of the first ratio and the weight factor of all pixels in each block is calculated to obtain the suppression degree of each block.
[0047] The formula for calculating the degree of suppression for any of the above blocks can be:
[0048] ;
[0049] In the formula, Indicates the first The degree of suppression of block segmentation; Indicates the first The total number of pixels in the block; Indicates the pixel ordinal number; Indicates the first The first block Local dark channel values of each pixel; Indicates the first The maximum value of the local dark channel value of all pixels in the block; Indicates the first The first block Local grayscale differences of individual pixels; To preset hyperparameters, in order to avoid Too large, leading to Convergence too fast, for example .
[0050] in, This represents the first ratio of any pixel in the block. Used for The smaller the normalization value, the more obvious the low grayscale area is, meaning the suppression level of the area is low and the possibility of defects is greater. In this case, the area needs to be enhanced significantly, that is, the suppression level of the block is reduced to highlight the weak contaminants. Conversely, the larger the value, the more obvious the suppression level of the area is, or the suppression level is low and there are no defects. In this case, the contrast of the current block needs to be suppressed to protect the current contrast from being destroyed by extreme peaks.
[0051] This represents the local grayscale difference of any pixel in the block. The smaller the value, the smaller the local difference of any pixel. Therefore, it is negatively correlated and normalized. The result of the negative correlation normalization is used as the weight factor of the first ratio of any pixel.
[0052] It should be noted that during image acquisition, the light source is positioned above the metal workpiece, and the workpiece surface is metal with high reflectivity. When the spray produces a large amount of water mist, the tiny droplets strongly scatter the light, and the scattered light directly enters the lens, forming a bright "white veil" (for example, when taking photos in foggy weather, the entire image appears white and the contrast is reduced). This means that the local dark channel value of pixels in the dense fog area is higher. When the water mist is very sparse or almost non-existent, there is less scattered light, and clean metal areas are brighter, but areas with rust, oil, pits, etc., remain darker. Therefore, in areas with relatively sparse water mist, if rust, oil, or pits are present, the dark channel value of pixels in that area is lower; if rust, oil, or pits are absent, the dark channel value is higher, but this does not affect the subsequent results. Therefore, areas with low suppression and no rust, oil, or pits are processed in the same way as areas with high suppression.
[0053] S3: Based on a preset time window, the real-time spray image is divided into multiple window time sequences. Each window time sequence contains multiple frames of spray images and is further divided into real-time blocks. The average suppression level of blocks at the same position in consecutive frames within any window time sequence is calculated. An initial limiting parameter and a reference suppression level are preset. The initial limiting parameter is corrected based on the difference between the reference suppression level and the average suppression level, resulting in the adjusted limiting parameter for each block.
[0054] It's important to note that directly below the spray area, when the water mist is at its densest, the image appears entirely white with extremely low contrast. However, numerous bright water droplet reflections appear in certain areas, which manifest as extremely high peaks on the histogram. A histogram is a tool used to statistically analyze the distribution of pixels at different gray levels in an image. The horizontal axis represents the image's gray levels. For an 8-bit grayscale image, the gray level range is from 0 (pure black) to 255 (pure white), totaling 256 levels. For color images, the gray level distribution of the RGB channels is typically analyzed separately. The vertical axis represents the number of pixels at the corresponding gray level, or pixel percentage. A higher value indicates a greater number of pixels at that gray level in the image. Using a large limiting parameter in this case can significantly amplify the water droplet reflection noise, resulting in noticeable blocky artifacts and grainy noise. Artifacts refer to false features in an image that do not exist on the actual surface of the workpiece. These features are not contaminants or defects of the workpiece itself, but are caused by interference or algorithm defects in the image acquisition and processing process. They can seriously interfere with the accurate identification of contaminants. The system is very likely to misjudge these artifacts as rust spots or oil stains, thus mistakenly believing that the workpiece is "severely contaminated", and then mistakenly increasing the spray pressure, resulting in unnecessary waste of resources.
[0055] At the edge of the spray area, the water mist is very thin, and the image regains normal contrast. At this time, the fine grease residue, thin rust, oxide layer and other contaminants on the workpiece surface appear as low-contrast dark areas. If the limiting parameters are too small, the histogram is hardly truncated (meaning a grayscale threshold range is set to force the grayscale values of pixels outside this range (such as extreme pixels that are too bright or too dark) to be limited within the threshold boundary, eliminating the interference of extreme grayscale values). The contrast improvement is very limited, and these truly weak attachments will be "submerged" in the background and cannot be detected, causing the system to mistakenly judge that it is "clean" when in fact it carries contaminants into the coating process, ultimately causing the coating to blister.
[0056] In one embodiment, the preset time window is 10 seconds. The collected spray images are divided based on the time window to obtain multiple window time sequences. For any given moment, the spray image collected at any given moment is taken as the last spray image data in its window time sequence, and spray image data of the preset time window length are obtained sequentially forward.
[0057] Obtain the average suppression level of any block in the spray images of consecutive frames in the window time sequence for the same position of the block.
[0058] Calculate the hyperbolic tangent function of the difference between the reference suppression level and the average suppression level of any block in the window timing, take half of it and calculate the sum with 1, which is taken as the first sum. The product of the first sum and the initial limiting parameter is taken as the adjustment limiting parameter of any block.
[0059] The formula for calculating the adjustment limit parameter of any of the above blocks can be:
[0060] ;
[0061] Indicates the first Adjust the amplitude limiting parameters in segments; This indicates the preset initial limiting parameters, for example... ; To preset the reference suppression level, for example ; Indicates the first The average degree of suppression of blocks within the time window.
[0062] For any given block, if the suppression level is high (dense fog area) or the suppression level is low but there are no defects (thin fog area but no pollutants), then these areas are considered high-brightness areas. In order to prevent the histograms corresponding to these areas from being destroyed by extreme high brightness values, it is necessary to protect the contrast of these areas. That is, it is necessary to appropriately reduce the limiting parameter to avoid excessive peak values of some gray levels. Conversely, it is necessary to appropriately increase the limiting parameter (thin fog area with pollutants) to improve the contrast of the block and enhance the visibility of weak pollutants.
[0063] S4: Process each segment based on the adjustment limit parameter to obtain the attachment density; preset the benchmark spray pressure and the attachment density reference value, and correct the benchmark spray pressure based on the difference between the attachment density and the attachment density reference value to obtain the adjustment spray pressure for adjusting the production line operation status.
[0064] Specifically, for each block, the histogram equalization algorithm is used to enhance the image of each block by adjusting the amplitude limiting parameter, resulting in an enhanced block image for each block.
[0065] The number of pixels of the attachment can be obtained by segmenting the enhanced block image using the Otsu thresholding algorithm; the ratio of the number of pixels in all enhanced block images to the total number of pixels in all enhanced block images is taken as the attachment density.
[0066] The hyperbolic tangent function of the difference between the density of the attached organism and the reference value of the attached organism density is calculated. After taking half of it, the sum of the result with 1 is calculated as the second sum. The product of the second sum and the reference spray pressure is used as the adjustment spray pressure.
[0067] The formula for calculating the adjustment of the spray pressure mentioned above can be:
[0068] ;
[0069] In the formula, This indicates an adjustment to the spray pressure; Indicates the preset reference spray pressure, for example ; This indicates the density of the currently detected deposits; Indicates a reference value for the density of the attached material, for example .
[0070] Preset suppression level, adjust spray pressure per Updated every second, but can be adjusted according to specific implementation requirements. For example, consider the time it takes for a metal workpiece to travel from the start of spraying to the area where the camera captures the image.
[0071] This completes the monitoring of the operating status of the metal workpiece pretreatment production line. By monitoring whether contaminants still exist on the workpiece after spraying, the spraying pressure can be adjusted in a timely manner.
[0072] This application also discloses a monitoring system for the operating status of a metal workpiece pretreatment production line, referring to... Figure 2 It includes a processor and a memory, the memory storing computer program instructions, which, when executed by the processor, implement a method for monitoring the operating status of a metal workpiece pretreatment production line according to this application.
[0073] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.
[0074] In this application, the aforementioned memory can be any tangible medium that contains or stores a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as resistive random access memory (DRAM), dynamic random access memory (DRAM), static random access memory (SRAM), etc., or any other medium that can be used to store desired information and can be accessed by an application program, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device.
[0075] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A method for monitoring the operating status of a metal workpiece pretreatment production line, characterized in that, include: Acquire a grayscale image of the metal workpiece surface and calculate the local grayscale difference of each pixel in the spray image within a preset neighborhood. The minimum gray value in the neighborhood of each pixel is taken as the local dark channel value of each pixel. The spray image is divided into multiple blocks by a grid. The suppression level of each block is calculated based on the local dark channel value and local grayscale difference of each pixel in each block, including: The ratio of the local dark channel value of any pixel in each block to the maximum local dark channel value of all pixels is used as the first ratio. Hyperparameters are set, and an exponential function is used to negatively correlate the product of the local gray level difference of any pixel with the hyperparameter, which is used as the weight factor of the first ratio. The average of the sum of the products of the first ratio and the weight factor of all pixels in each block is calculated to obtain the suppression degree of each block. The real-time spray images are divided into multiple window time sequences based on a preset time window. Each window time sequence contains multiple frames of spray images and is further divided into real-time blocks. The average suppression level of blocks at the same position in consecutive frames within any window time sequence is calculated. An initial limiting parameter and a reference suppression level are preset. The initial limiting parameter is corrected based on the difference between the reference suppression level and the average suppression level, resulting in the adjusted limiting parameter for each block. Each segment is processed based on the adjustment limit parameter to obtain the attachment density; a preset benchmark spray pressure and attachment density reference value are used to correct the benchmark spray pressure based on the difference between the attachment density and the attachment density reference value, and the adjusted spray pressure is obtained to adjust the production line operation status.
2. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for calculating the local grayscale difference is as follows: calculate the mean of the sum of the squared differences between the grayscale values of each pixel and all pixels in the neighborhood, and use this as the grayscale difference. After taking the square root of the grayscale difference, standard normalize it to obtain the local grayscale difference.
3. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for dividing the spray image into multiple blocks by grid is as follows: if the spray image is less than the size of a block, it is filled to the size of a block.
4. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for dividing the window time sequence is as follows: for any given moment, the spray image acquired at any given moment is taken as the last spray image data in the window time sequence of any given moment, and spray image data of a preset time window length are acquired sequentially forward.
5. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for calculating the adjustment limiting parameter of each block is as follows: calculate the hyperbolic tangent function of the difference between the reference suppression degree and the average suppression degree of each block in the window timing, take half of it and calculate the sum with 1, which is taken as the first sum. The product of the first sum and the initial limiting parameter is taken as the adjustment limiting parameter of each block.
6. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for calculating the density of the attachment is as follows: based on adjusting the amplitude limiting parameter, the histogram equalization algorithm is used to enhance the image of each block to obtain the enhanced block image of each block. The Otsu threshold segmentation algorithm is used to segment each enhanced block image to obtain the number of pixels of the attachment in each enhanced block image. The attachment density is the ratio of the total number of pixels of the attachments in all enhanced block images to the total number of pixels in all enhanced block images.
7. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, The method for calculating the adjusted spray pressure is as follows: calculate the hyperbolic tangent function of the difference between the density of the attached material and the reference value of the attached material density, take half of it and calculate the sum with 1, which is taken as the second sum. The product of the second sum and the reference spray pressure is taken as the adjusted spray pressure.
8. The method for monitoring the operating status of a metal workpiece pretreatment production line according to claim 1, characterized in that, Also includes: The level of suppression and the adjustment of spray pressure are updated every time window.
9. A monitoring system for the operating status of a metal workpiece pretreatment production line, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a method for monitoring the operating status of a metal workpiece pretreatment production line according to any one of claims 1-8.