A Visual Inspection System for the Surface Quality of Sandblasted Parts

Abstract

The present invention discloses a visual inspection system for the surface quality of sandblasted parts, which relates to the technical field of sandblasting quality inspection and realizes the dynamic optimization of the sandblasting process. According to the generation sequence of multi-spectral sandblasting image data, the present invention sequentially compares each multi-spectral sandblasting image data with the pixels in the subsequent generated multi-spectral sandblasting image data, classifies the pixels in the multi-spectral sandblasting image data into product appearance pixels and sandblasting result pixels according to the comparison results, and then establishes a standard product sandblasting result image model based on the product appearance pixels and sandblasting result pixels. At the same time, a three-dimensional point cloud distribution is set in the standard product sandblasting result image model according to the sandblasting work event, and the three-dimensional point cloud distribution is compared with the local three-dimensional point position distribution map in chronological order, and corresponding spray gun adjustment decisions are generated and executed according to the comparison results.

Description

Technical Field

[0001] The present invention relates to the technical field of sandblasting quality detection, and specifically to a visual detection system for the surface quality of sandblasted parts. Background Art

[0002] In modern manufacturing, sandblasting is a common surface treatment process used to improve the surface quality and appearance of parts. However, traditional sandblasting processes often rely on manual inspection to evaluate the surface quality of sandblasted parts. This method is not only time-consuming and laborious, but also greatly affected by subjective factors, making it difficult to ensure quality consistency.

[0003] Existing technologies usually use cameras to capture images during the sandblasting process and identify surface defects through image processing algorithms. However, these systems often have problems such as low detection accuracy and poor adaptability, especially when faced with complex sandblasting work events and variable sandblasting results, and their performance is particularly obvious. Therefore, a visual detection system for the surface quality of sandblasted parts is provided. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a visual detection system for the surface quality of sandblasted parts.

[0005] In order to achieve the above purpose, the present invention provides the following technical solutions:

[0006] A visual detection system for the surface quality of sandblasted parts, including a cloud computing platform, which is communicatively connected to a sandblasting video acquisition module, a standard part feature analysis module, and a characterization defect detection module;

[0007] The sandblasting video acquisition module is used to set several sandblasting work events, and then through a gravity sensing device and a multispectral imager, collect multispectral sandblasting video data and overall displacement curves when various sandblasting work events are executed several times, and collect real-time laser video data and real-time overall displacement curves during the execution of real-time sandblasting work events through a laser imager and a gravity sensing device;

[0008] The standard part feature analysis module is used to establish a three-dimensional coordinate system, and divide the multispectral sandblasting video data into several pieces of multispectral sandblasting image data frame by frame, and then map the multispectral sandblasting image data onto the three-dimensional coordinate system in sequence according to the overall displacement curve;

[0009] In the order of generating multi - spectral sandblasting image data, each multi - spectral sandblasting image data is compared with the pixels in the subsequently generated multi - spectral sandblasting image data in sequence. According to the comparison results, the pixels in the multi - spectral sandblasting image data are divided into product appearance pixels and sandblasting result pixels. Furthermore, a standard product sandblasting result image model is established based on the product appearance pixels and sandblasting result pixels. At the same time, a three - dimensional point cloud distribution is set in the standard product sandblasting result image model according to the sandblasting work event;

[0010] The characterization defect detection module is used to generate a local three - dimensional point position distribution map based on the real - time laser video data and the real - time overall displacement curve, and map it onto the standard product sandblasting result image model. The local three - dimensional point position distribution map is compared with the three - dimensional point cloud distribution in chronological order, and corresponding spray gun adjustment decisions are generated and executed according to the comparison results.

[0011] Furthermore, the acquisition processes of the multi - spectral sandblasting video data, the overall displacement curve, and the real - time laser video data include:

[0012] The sandblasting video acquisition module pre - stores product sandblasting plans and is communicatively connected to n sandblasting machines. The product sandblasting plans include the sandblasted product name, sandblasting pattern, and expected sandblasting thickness;

[0013] Among them, the sandblasting machine is equipped with a multi - spectral imager, a laser imager, a gravity sensing device, and a spray gun, and n is a natural number greater than 0;

[0014] According to the product sandblasting plan, several sandblasting work events are set, and each sandblasting work event is executed m times. During the execution of each sandblasting work event, the multi - spectral imager real - time collects multi - spectral sandblasting video data during the interaction between the spray gun and the outer surface of the target product. At the same time, when the spray gun makes an overall displacement, the gravity sensing device records the overall displacement curve of the spray gun, where m is a natural number greater than 0;

[0015] During the execution of the real - time sandblasting work event by each sandblasting machine, real - time laser video data during the interaction between the spray gun and the outer surface of the target product is collected, and n corresponding real - time overall displacement curves are synchronously collected.

[0016] Furthermore, the process of mapping the multi - spectral sandblasting image data onto the three - dimensional coordinate system according to the overall displacement curve includes:

[0017] The multi - spectral sandblasting video data is segmented into several pieces of multi - spectral sandblasting image data frame by frame. A three - dimensional coordinate system is established, and the overall displacement curves generated by each sandblasting machine in the same sandblasting work event and the same execution times are matched and spliced. Then, according to the change trend of each overall displacement curve at each time node, the multi - spectral sandblasting image data generated by each sandblasting machine is overlapped and mapped into the three - dimensional coordinate system.

[0018] Further, the process of classifying pixels in the multispectral sandblasting image data into product appearance pixels and sandblasting result pixels includes:

[0019] Obtain the pixel values of each pixel in each piece of multispectral sandblasting image data, and starting from the multispectral sandblasting image data generated at the first time node, according to the overlapping direction of the multispectral sandblasting image data, extract the image segments that partially overlap with the multispectral sandblasting image data generated at the first time node from the k multispectral sandblasting image data after the multispectral sandblasting image data generated at the first time node. Denote it as the overlapping image segment, and k takes any integer between (0, 10);

[0020] Set the first pixel difference threshold and the second pixel difference threshold, and judge the magnitude relationship between the pixel differences between pixels in the overlapping image segments of the k multispectral sandblasting image data after the multispectral sandblasting image data generated at each time node and the first pixel difference threshold and the second pixel difference threshold;

[0021] Label each pixel as a product appearance pixel and a sandblasting result pixel according to the judgment result, and classify the product appearance pixels into textured product appearance pixels and flat product appearance pixels.

[0022] Further, the process of establishing the standard product sandblasting result image model includes:

[0023] Statistically analyze the product sandblasting result image models generated corresponding to the same sandblasting work event but different execution times, count the occurrence times of the image model parts containing sandblasting result pixels at each position, and set a frequency threshold;

[0024] Convert the image model parts of the sandblasting result pixels with occurrence times less than the frequency threshold into the image model parts of the corresponding product appearance pixels, otherwise do nothing;

[0025] When all the judgment conversions of the image model parts corresponding to the sandblasting result pixels are completed, obtain the standard product sandblasting result image model, and then set the texture curve according to the textured product appearance pixels in the standard product sandblasting result image model.

[0026] Further, the process of setting the three-dimensional point cloud distribution in the standard product sandblasting result image model includes:

[0027] Furthermore, according to the reflection degree of the laser signal by the sandblasting material and the expected sandblasting thickness, set the three-dimensional point cloud distribution on the image model parts corresponding to the sandblasting result pixels in the standard product sandblasting result image model, and set the horizontal point cloud height. Then, label the relative sandblasting height for each three-dimensional point cloud distribution according to the horizontal point cloud height and the texture curve distribution.

[0028] Further, the process of comparing the local three-dimensional point distribution map with the three-dimensional point cloud distribution includes:

[0029] Dividing the real-time laser video data into several pieces of real-time laser image data frame by frame, and sequentially overlapping and distributing the real-time laser image data in the three-dimensional coordinate system according to the changing trends of each real-time overall displacement curve;

[0030] Generating corresponding local three-dimensional point distribution maps based on the real-time laser image data generated at each time node, respectively mapping each local three-dimensional point distribution map onto the standard product sandblasting result image model, setting the point cloud height difference threshold, and sequentially comparing the total point cloud height difference between the real-time point cloud height and the corresponding relative sandblasting height of the local three-dimensional point distribution maps generated at each time node;

[0031] And determining whether there is an overlap in the part of the model corresponding to the product appearance pixels in the standard product sandblasting result image model for the local three-dimensional point distribution maps generated at each time node, and judging whether the sandblasting operation of the corresponding sandblaster meets the sandblasting requirements at the current time node according to the judgment result.

[0032] Further, the generation process of the spray gun adjustment decision includes:

[0033] If it is judged that the reason is that the total point cloud height difference is greater than the point cloud height difference threshold, and half of the real-time point cloud heights are greater than the corresponding relative sandblasting height, or half of the real-time point cloud heights are less than or equal to the corresponding relative sandblasting height, then the spray gun adjustment decision is used to increase or decrease the sandblasting intensity of the corresponding spray gun at the next time node;

[0034] If it is judged that the reason is that there is an overlap in the part of the model corresponding to the product appearance pixels in the standard product sandblasting result image model for the local three-dimensional point distribution maps generated at each time node, then generate a spray gun adjustment decision according to the opposite direction of the relative direction of the existing part;

[0035] Send the spray gun adjustment decision to the corresponding sandblaster, and then the sandblaster adjusts the sandblasting intensity or the sandblasting direction according to the spray gun adjustment decision.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] 1. The present invention divides the multi-spectral sandblasting video data into multi-spectral sandblasting image data frame by frame, and then maps these image data onto the three-dimensional coordinate system according to the overall displacement curve. By sequentially comparing the multi-spectral sandblasting image data, the product appearance pixels and the sandblasting result pixels are distinguished, and a standard product sandblasting result image model is established. At the same time, a three-dimensional point cloud distribution is set on the model, thereby realizing accurate capture of the surface details of the sandblasted parts and providing a data basis for the accuracy of subsequent sandblasting defect detection.

[0038] 2. The present invention generates a local three-dimensional point distribution map based on real-time laser video data and real-time overall displacement curve, maps it onto the standard product sandblasting result image model, compares the local three-dimensional point distribution map with the three-dimensional point cloud distribution in chronological order, and based on the real-time collected data and comparison results, immediately generates a decision on spray gun adjustment, realizing the dynamic optimization of the sandblasting process while improving the quality of sandblasted products. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific embodiments, structures, features and their effects of the present invention as follows.

[0041] As Figure 1 shown, a visual inspection system for the surface quality of sandblasted parts includes a cloud computing platform, and the cloud computing platform is communicatively connected to a sandblasting video acquisition module, a standard part feature analysis module, and a characterization defect detection module;

[0042] The sandblasting video acquisition module is used to set several sandblasting work events, and then through a gravity sensing device and a multispectral imager, collect multispectral sandblasting video data and overall displacement curve when various sandblasting work events are executed several times, and collect real-time laser video data and real-time overall displacement curve during the execution of real-time sandblasting work events through a laser imager and a gravity sensing device;

[0043] The standard part feature analysis module is used to establish a three-dimensional coordinate system, divide the multispectral sandblasting video data into several pieces of multispectral sandblasting image data frame by frame, and then map the multispectral sandblasting image data onto the three-dimensional coordinate system in sequence according to the overall displacement curve;

[0044] In the order of generation of the multispectral sandblasting image data, compare the pixels in each multispectral sandblasting image data with the pixels in the subsequent generated multispectral sandblasting image data in sequence. According to the comparison results, divide the pixels in the multispectral sandblasting image data into product appearance pixels and sandblasting result pixels, and then establish a standard product sandblasting result image model based on the product appearance pixels and sandblasting result pixels. At the same time, set a three-dimensional point cloud distribution in the standard product sandblasting result image model according to the sandblasting work event;

[0045] The characterization defect detection module is used to generate a local three-dimensional point distribution map based on the real-time laser video data and real-time overall displacement curve, map it onto the standard product sandblasting result image model, compare the local three-dimensional point distribution map with the three-dimensional point cloud distribution in chronological order, and generate and execute a corresponding spray gun adjustment decision according to the comparison results.

[0046] Further, the working principle of the present invention will be specifically described below through embodiments:

[0047] The staff uploads several product sandblasting schemes to the sandblasting video acquisition module and the standard part feature analysis module. The product sandblasting schemes include the name of the sandblasted product, the sandblasting pattern, and the expected sandblasting thickness;

[0048] The sandblasting video acquisition module is communicatively connected to n sandblasting machines, where the sandblasting machines are provided with multi-spectral imagers, laser imagers, gravity sensing devices, and spray guns, and n is a natural number greater than 0;

[0049] Numbers a1, a2, a3,..., a n are set for each sandblasting machine, and several sandblasting work events are set according to the product sandblasting scheme. Any one of the sandblasted product names or sandblasting patterns corresponding to different sandblasting work events is different;

[0050] Each sandblasting work event is executed m times. During the execution of each sandblasting work event, the spray gun of the sandblasting machine is adjusted according to the sandblasting pattern to perform sandblasting operations on the outer surface of the target product. At the same time, the multi-spectral imager carried by the sandblasting machine collects multi-spectral sandblasting video data during the interaction between the spray gun and the outer surface of the target product in real time. At the same time, when the spray gun makes an overall displacement, the gravity sensing device records the overall displacement curve of the spray gun, where m is a natural number greater than 0;

[0051] It should be noted that the video acquisition directions of the multi-spectral imager and the laser imager are always consistent with the sandblasting operation direction of the spray gun associated with them;

[0052] During the execution of the real-time sandblasting work event by each sandblasting machine, the laser imager on the sandblasting machine collects real-time laser video data during the interaction between the spray gun and the outer surface of the target product in the same manner as the multi-spectral imager, and synchronously collects n corresponding real-time overall displacement curves.

[0053] Further, after all sandblasting work events are executed m times, the sandblasting video data acquisition module will label the corresponding sandblasting machine numbers and sandblasting work events for each multi-spectral sandblasting video data and overall displacement curve, and then send all multi-spectral sandblasting video data and overall displacement curves to the standard part feature analysis module, and send the real-time laser video data and real-time overall displacement curves to the characterization defect detection module;

[0054] The standard part feature analysis module divides the multi-spectral sandblasting video data into several pieces of multi-spectral sandblasting image data frame by frame. Since the shooting angle of the multi-spectral imager is always consistent with the spray gun, the outer surface position of the target product is always included within the shooting range of the multi-spectral imager. At the same time, due to the different spectral absorption rates of the same frequency band before and after sandblasting on the outer surface of the target product, there are obvious differences in the multi-spectral sandblasting image data generated at the corresponding positions at the two time nodes before and after;

[0055] Establish a three-dimensional coordinate system, and match and splice the overall displacement curves generated by each sandblaster during the same sandblasting work event and the same execution times;

[0056] Furthermore, according to the change trend of each overall displacement curve at each time node, the multi-spectral sandblasting image data generated by each sandblaster is overlapped and mapped in the three-dimensional coordinate system, where the change trend of each multi-spectral sandblasting image data in the three-dimensional coordinate system is the same as that of the overall displacement curve;

[0057] Obtain the pixel values of each pixel in each piece of multi-spectral sandblasting image data, and starting from the multi-spectral sandblasting image data generated at the first time node, according to the overlapping direction of the multi-spectral sandblasting image data, extract the image fragments that partially overlap with the multi-spectral sandblasting image data generated at the first time node from the k multi-spectral sandblasting image data after the multi-spectral sandblasting image data generated at the first time node, denoted as overlapping image fragments, where k takes any integer between (0, 10);

[0058] It should be noted that since there are generally some uneven texture parts on the outer surface of the target product, such as the groove curve on the vehicle front cover, during the process of the multi-spectral imager collecting images of the corresponding parts, due to the lighting reason, the collection effect of the multi-spectral imager on the texture groove part is darker than that on the texture protrusion part, resulting in differences in the pixel values of the corresponding parts, but the difference amplitude of the pixel values is smaller than that before and after sandblasting;

[0059] Set a first pixel difference threshold and a second pixel difference threshold, where the first pixel difference threshold is less than the second pixel difference threshold, and determine whether the pixel difference between each pixel in each overlapping image fragment and the corresponding part pixel in the multi-spectral sandblasting image data generated at the first time node is greater than or equal to the second pixel difference threshold;

[0060] If the pixel difference is greater than or equal to the second pixel difference threshold, mark the corresponding pixel in the multi-spectral sandblasting image data generated at the first time node as a sandblasting result pixel. If the pixel difference is less than the second pixel difference threshold, mark the corresponding pixel as a product appearance pixel;

[0061] Select the k multi-spectral sandblasting image data after the multi-spectral sandblasting image data generated at the second time node, and repeat the process of annotating each pixel in the multi-spectral sandblasting image data generated at the first time node, and so on, until the pixels in the multi-spectral sandblasting image data generated at the last time node are all annotated and then stop;

[0062] Adopt the process of annotating the pixels of the sandblasting result, and judge again the size relationship between the pixel difference between the product appearance pixels and the first pixel difference threshold in the overlapping image segments of the k multi-spectral sandblasting image data after the multi-spectral sandblasting image data generated at each time node, so as to classify the product appearance pixels into texture product appearance pixels and flat product appearance pixels.

[0063] Furthermore, according to the distribution of the sandblasting result pixels and the product appearance pixels in each multi-spectral sandblasting image data under the same execution times, as well as the distribution and overlapping conditions of each multi-spectral sandblasting image data in the three-dimensional coordinate system, splice the product appearance pixels and the sandblasting result pixels in each multi-spectral sandblasting image data respectively, so as to obtain the product appearance image model and the product sandblasting result image model before and after the sandblasting work event of the target product under the corresponding execution times. For the product sandblasting result image model, according to the distribution of the texture product appearance pixels, texture curves are set on the surface of the product sandblasting result image model;

[0064] It should be noted that the overall structures of the product appearance image model and the product sandblasting result image model are the same, but the product sandblasting result image model simultaneously includes the image model parts corresponding to the sandblasting result pixels and the product appearance pixels, while the product appearance image model is only generated according to the product appearance pixels;

[0065] Count the occurrence times of the image model parts containing the sandblasting result pixels at each position in the product sandblasting result image models generated corresponding to the same sandblasting work event but different execution times, and set a frequency threshold, where the frequency threshold is less than m;

[0066] Convert the image model parts of the sandblasting result pixels with the occurrence times less than the frequency threshold into the image model parts of the corresponding product appearance pixels, otherwise do nothing;

[0067] When all the determination and conversion of the image model parts corresponding to the sandblasting result pixels are completed, a standard product sandblasting result image model is obtained, and then the texture curves on the surface of the product sandblasting result image model are mapped onto the standard product sandblasting result image model;

[0068] Since during the process of sandblasting on the outer surface of the target product, as the sandblasting thickness changes continuously, the reflection degree of the laser signal emitted by the laser imager on the outer surface of the target product also changes accordingly;

[0069] Furthermore, according to the reflection degree of the laser signal by the sandblasting material and the desired sandblasting thickness, a three-dimensional point cloud distribution is set for the image model part corresponding to the sandblasting result pixels on the standard product sandblasting result image model, and the horizontal point cloud height is set. Then, according to the horizontal point cloud height and the texture curve distribution, the relative sandblasting height is marked for each three-dimensional point cloud distribution.

[0070] Further, when the defect detection module receives the real-time laser video data and the real-time overall displacement curve from the sandblasting image acquisition module, the real-time laser video data is divided into several pieces of real-time laser image data frame by frame;

[0071] In the process of mapping the multi-spectral sandblasting image data to the three-dimensional coordinate system, according to the change trend of each real-time overall displacement curve, the real-time laser image data is sequentially overlapped and distributed in the three-dimensional coordinate system, and according to the subsequent generated real-time laser video data and real-time overall displacement curve, the distribution of the real-time laser image data in the three-dimensional coordinate system is updated in real time;

[0072] At the same time, since the reflection degrees of the laser signals by the sandblasting results with different thicknesses are different, local three-dimensional point position distribution maps are generated according to the real-time laser image data generated at each time node, and each local three-dimensional point position distribution map is respectively mapped onto the standard product sandblasting result image model;

[0073] A point cloud height difference threshold is set, and the total point cloud height difference between the real-time point cloud height of each local three-dimensional point position distribution map generated at each time node and the corresponding relative sandblasting height is compared in sequence;

[0074] If the total point cloud height difference is less than or equal to the point cloud height difference threshold, it is determined that the sandblasting operation of the corresponding sandblasting machine at the current time node meets the sandblasting requirements;

[0075] If the total point cloud height difference is greater than the point cloud height difference threshold, it is determined that the sandblasting operation of the corresponding sandblasting machine at the current time node does not meet the sandblasting requirements;

[0076] At the same time, it is judged whether there is an overlap in the model part corresponding to the product appearance pixels in the standard product sandblasting result image model for each local three-dimensional point position distribution map generated at each time node. If so, it is determined that the sandblasting operation of the corresponding sandblasting machine at the current time node does not meet the sandblasting requirements;

[0077] If not, it is determined that the sandblasting operation of the corresponding sandblasting machine at the current time node meets the sandblasting requirements.

[0078] Further, if it is determined that the sandblasting operation of the sandblasting machine at the current time node meets the sandblasting requirements, no operation is performed;

[0079] If it is determined that the sandblasting operation of the sandblasting machine at the current time node does not meet the sandblasting requirements, then according to the reason for the judgment, a spray gun adjustment decision is set for the corresponding sandblasting machine;

[0080] If it is determined that the reason is that the total value of the point cloud height difference is greater than the point cloud height difference threshold, and half of the real-time point cloud height is greater than the corresponding relative sandblasting height, or half of the real-time point cloud height is less than or equal to the corresponding relative sandblasting height, then the spray gun adjustment decision is used to increase or decrease the sandblasting intensity of the corresponding spray gun at the next time node;

[0081] If it is determined that the reason is that there is an overlap in the part of the product appearance pixel corresponding model in the standard product sandblasting result image model in the local three-dimensional point position distribution maps generated at each time node, then a spray gun adjustment decision is generated according to the opposite direction of the relative direction of the existing part;

[0082] Send the spray gun adjustment decision to the corresponding sandblasting machine, and then the sandblasting machine adjusts the sandblasting intensity or the sandblasting direction according to the spray gun adjustment decision;

[0083] Repeat the above process of generating the spray gun adjustment decision until the corresponding real-time sandblasting work event ends.

[0084] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in any form. Although the present invention has been disclosed as above with the preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the equivalent embodiments with equivalent changes within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A visual inspection system for the surface quality of sandblasted parts, including a cloud computing platform, characterized in that, The cloud computing platform is communicatively connected to a sandblasting video acquisition module, a standard part feature analysis module, and a characterization defect detection module; The sandblasting video acquisition module is used to set several sandblasting work events, and then, through a gravity sensing device and a multi-spectral imager, collect multi-spectral sandblasting video data and an overall displacement curve when each sandblasting work event is executed several times, and collect real-time laser video data and a real-time overall displacement curve during the execution of a real-time sandblasting work event through a laser imager and a gravity sensing device; The acquisition process of the multi-spectral sandblasting video data, the overall displacement curve, and the real-time laser video data includes: The sandblasting video acquisition module pre-stores a product sandblasting plan and is communicatively connected to n sandblasting machines. The product sandblasting plan includes the name of the sandblasted product, the sandblasting pattern, and the expected sandblasting thickness; Among them, the sandblasting machine is provided with a multi-spectral imager, a laser imager, a gravity sensing device, and a spray gun, and n is a natural number greater than 0; Set several sandblasting work events according to the product sandblasting plan, and execute each sandblasting work event m times. During the execution of each sandblasting work event, the multi-spectral imager collects multi-spectral sandblasting video data during the interaction between the spray gun and the outer surface of the target product in real time. At the same time, when the spray gun makes an overall displacement, the gravity sensing device records the overall displacement curve of the spray gun, where m is a natural number greater than 0; During the execution of the real-time sandblasting work event by each sandblasting machine, collect real-time laser video data during the interaction between the spray gun and the outer surface of the target product, and synchronously collect the corresponding real-time overall displacement curve; The process of sequentially mapping the multi-spectral sandblasting image data onto a three-dimensional coordinate system according to the overall displacement curve includes: Segment the multi-spectral sandblasting video data into several pieces of multi-spectral sandblasting image data frame by frame, establish a three-dimensional coordinate system, match and splice the overall displacement curves generated by each sandblasting machine in the same sandblasting work event and the same execution times, and then, according to the change trend of each overall displacement curve at each time node, overlap and map each multi-spectral sandblasting image data onto the three-dimensional coordinate system; The standard part feature analysis module is used to establish a three-dimensional coordinate system, segment the multi-spectral sandblasting video data into several pieces of multi-spectral sandblasting image data frame by frame, and then, according to the overall displacement curve, sequentially map the multi-spectral sandblasting image data onto the three-dimensional coordinate system; Sequentially compare the pixels in each multi-spectral sandblasting image data with the pixels in the subsequent generated multi-spectral sandblasting image data according to the generation order of the multi-spectral sandblasting image data. Divide the pixels in the multi-spectral sandblasting image data into product appearance pixels and sandblasting result pixels according to the comparison result. Then, establish a standard product sandblasting result image model according to the product appearance pixels and the sandblasting result pixels, and at the same time, set a three-dimensional point cloud distribution in the standard product sandblasting result image model according to the sandblasting work event; The process of dividing the pixels in the multi-spectral sandblasting image data into product appearance pixels and sandblasting result pixels includes: Obtain the pixel values of each pixel in each piece of multispectral sandblasting image data, and starting from the multispectral sandblasting image data generated at the first time node, extract the image segments that partially overlap with the multispectral sandblasting image data generated at the first time node from the k multispectral sandblasting image data after the multispectral sandblasting image data generated at the first time node. Denote it as the overlapping image segment, where k takes any integer between (0, 10); Furthermore, set the first pixel difference threshold and the second pixel difference threshold, and sequentially judge the magnitude relationship between the pixel differences between pixels in the overlapping image segments of the k multispectral sandblasting image data after the multispectral sandblasting image data generated at each time node and the first pixel difference threshold and the second pixel difference threshold; According to the judgment results, label each pixel as a product appearance pixel and a sandblasting result pixel, and divide the product appearance pixels into texture product appearance pixels and flat product appearance pixels; The establishment process of the standard product sandblasting result image model includes: Statistically analyze the product sandblasting result image models generated corresponding to the same sandblasting work event but different execution times, count the occurrence times of the image model parts containing sandblasting result pixels at each position, and set a frequency threshold. Convert the image model parts of the sandblasting result pixels with occurrence times less than the frequency threshold into the image model parts of the corresponding product appearance pixels, otherwise do nothing; When all the judgment and conversion of the image model parts corresponding to the sandblasting result pixels are completed, obtain the standard product sandblasting result image model, and set texture curves on the standard product sandblasting result image model according to the texture product appearance pixels; The process of setting the three-dimensional point cloud distribution on the standard product sandblasting result image model includes: According to the reflection degree of the laser signal by the sandblasting material and the expected sandblasting thickness, set the three-dimensional point cloud distribution on the image model parts corresponding to the sandblasting result pixels on the standard product sandblasting result image model, and set the horizontal point cloud height. Then, label the relative sandblasting height for each three-dimensional point cloud distribution according to the horizontal point cloud height and the texture curve distribution; The characterization defect detection module is used to generate a local three-dimensional point position distribution map based on the real-time laser video data and the real-time overall displacement curve, map it onto the standard product sandblasting result image model, sequentially compare the local three-dimensional point position distribution maps with the three-dimensional point cloud distribution in chronological order, and generate and execute corresponding spray gun adjustment decisions according to the comparison results; The process of comparing the local three-dimensional point position distribution map with the three-dimensional point cloud distribution includes: Divide the real-time laser video data into several pieces of real-time laser image data frame by frame, and according to the change trend of each real-time overall displacement curve, sequentially overlap and distribute the real-time laser image data in a three-dimensional coordinate system; Generate corresponding local three-dimensional point position distribution maps based on the real-time laser image data generated at each time node, map each local three-dimensional point position distribution map onto the standard product sandblasting result image model respectively, set a point cloud height difference threshold, and sequentially compare the total point cloud height difference between the real-time point cloud height of the local three-dimensional point position distribution maps generated at each time node and the corresponding relative sandblasting height; And determine whether there is an overlap in the part of the model corresponding to the product appearance pixels in the standard product sandblasting result image model for the local three-dimensional point position distribution maps generated at each time node, and judge whether the sandblasting operation of the corresponding sandblaster at the current time node meets the sandblasting requirements according to the judgment result; The generation process of the spray gun adjustment decision includes: If the judgment reason is that the total value of the point cloud height difference is greater than the point cloud height difference threshold, and half of the real-time point cloud height is greater than the corresponding relative sandblasting height, or half of the real-time point cloud height is less than or equal to the corresponding relative sandblasting height, the spray gun adjustment decision is used to increase or decrease the sandblasting intensity of the corresponding spray gun at the next time node; If the judgment reason is that there is an overlap in the part of the model corresponding to the product appearance pixels in the standard product sandblasting result image model for the local three-dimensional point position distribution maps generated at each time node, the spray gun adjustment decision is generated according to the opposite direction of the relative direction of the existing part.

Images