Waterproof coiled material performance evaluation system based on machine vision

Abstract

The invention relates to the technical field of waterproof coiled materials, in particular to a waterproof coiled material performance evaluation system based on machine vision, and the system comprises a control terminal which is a main control terminal of the system and is used for sending out an execution command; the acquisition module is used for acquiring image data of the waterproof roll; the segmentation module is used for receiving the waterproof roll image data acquired by the acquisition module and performing segmentation processing on the waterproof roll image data; the conversion module is used for continuously receiving the sub-waterproof roll image data obtained by segmentation in the segmentation module and converting the sub-waterproof roll image data into gray level images; the performance of the waterproof coiled material is judged through image acquisition, analysis and evaluation of the waterproof coiled material, manual detection is replaced to a large extent, the production efficiency and the qualified rate of the waterproof coiled material are improved, meanwhile, the labor cost in the waterproof coiled material detection stage in the waterproof coiled material production process is reduced, and the production efficiency of the waterproof coiled material is improved. And the intelligence of the waterproof roll production stage in the waterproof roll industry is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of waterproof coiled materials, and in particular to a waterproof coiled material performance evaluation system based on machine vision. Background Art

[0002] Waterproof membrane is an important building waterproof material, commonly used in waterproofing treatment of roofs, basements, tunnels and other projects.

[0003] It is usually made of polymer materials or asphalt, and has excellent waterproof performance. Its material is tough and can effectively resist the penetration of rainwater, groundwater, etc. The waterproof membrane is easy to construct and can be cut and laid according to the shape of the construction site. During laying, the membranes are tightly connected through professional bonding or welding processes to form a complete waterproof layer.

[0004] The invention patent with application number 202211002950.9 discloses a waterproof cloth performance detection method based on image recognition, which is characterized in that the method includes the following steps: respectively obtaining a grayscale image of the surface of the waterproof cloth to be detected and a grayscale image of the surface of the waterproof cloth to be detected after being sprinkled with water, recorded as a first detection image and a second detection image; respectively calculating the average grayscale value of the first detection image and the second detection image, and calculating the grayscale difference degree according to the difference in the average grayscale value of the first detection image and the second detection image; segmenting the second detection image to obtain a water droplet-containing area, and for each water droplet-containing area, obtaining the grayscale difference of all adjacent edge pixels of the water droplet-containing area. The angle between the connecting line between the two regions and the horizontal line is calculated, and the proximity of adjacent edge pixel points is calculated according to the difference between the angles; if the proximity is less than the set threshold, the edge pixel point corresponding to the proximity is obtained and recorded as an abnormal pixel point, and then the number of all abnormal pixel points in the area containing water droplets is obtained; the ratio of the number of smooth and flat pixel points on the edge of the area containing water droplets to the total number of edge pixel points in the area containing water droplets is calculated, and recorded as the smoothness of the area containing water droplets; the average value of the smoothness of all areas containing water droplets is obtained, and recorded as the average smoothness of the areas containing water droplets, and the first performance evaluation index is obtained according to the grayscale difference degree and the average smoothness of the areas containing water droplets.

[0005] This application aims to solve the problem that "when conducting waterproof testing on waterproof cloth, water is mostly poured directly on the cloth, and then the waterproof effect is judged by touch. Such manual testing is easily affected by subjective factors and other external factors, and the testing efficiency is low, and it is impossible to accurately judge the waterproof performance of the cloth."

[0006] At present, after the production of waterproof membranes, they are directly put into use after only a simple manual visual inspection. Some minor defects on the surface of the finished waterproof membranes cannot be detected, so that after being put into use, their waterproof performance is poor and their life attenuation cycle is short.

[0007] To this end, we proposed a waterproof membrane performance evaluation system based on machine vision. Summary of the invention

[0008] In view of the above-mentioned shortcomings of the prior art, the present invention provides a waterproof membrane performance evaluation system based on machine vision, which solves the technical problems raised in the above-mentioned background technology.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] A waterproofing membrane performance evaluation system based on machine vision, comprising:

[0011] The control terminal is the main control end of the system and is used to issue execution commands; the acquisition module is used to acquire the image data of the waterproofing membrane; the segmentation module is used to receive the image data of the waterproofing membrane acquired by the acquisition module and perform segmentation processing on the image data of the waterproofing membrane; the conversion module is used to continuously receive the sub-waterproofing membrane image data obtained by the segmentation module and convert the sub-waterproofing membrane image data into grayscale images; the evaluation module is used to obtain all grayscale images converted by the conversion module and evaluate the comprehensive similarity between all grayscale images; the judgment module is used to receive the comprehensive similarity between all grayscale images and judge whether the waterproofing membrane image data source is qualified based on the comprehensive similarity.

[0012] Furthermore, the acquisition module is provided with submodules at the lower level, including:

[0013] A storage unit, used to receive the waterproofing coiled material image data collected by the collection module and store the waterproofing coiled material image data;

[0014] An optimization unit, used for selecting the waterproofing coiled material image data stored in the storage unit and optimizing the image data;

[0015] Among them, the acquisition module is integrated by an industrial camera with a resolution of 50 million pixels. When the optimization unit selects the waterproof coil material image data stored in the storage unit, it selects in sequence based on the storage sequence of the waterproof coil material image data, selects a group of waterproof coil material image data each time, and after the selected waterproof coil material image data is optimized, it synchronously iterates to the corresponding original waterproof coil material image data storage position in the storage unit.

[0016] Furthermore, the optimization processing logic of the waterproofing coiled material image data in the optimization unit is expressed as:

[0017]

[0018] Where: P(x, y) is the optimized image pixel value; P0(x, y) is the original image pixel value; k is the adjustment parameter; G x (x, y) is the horizontal gradient of the image; G y (x,y) is the vertical gradient of the image;

[0019] Among them, when the optimization unit optimizes the waterproof roll material image data, it first divides the waterproof roll material image data into equal parts to obtain several groups of sub-waterproof roll material image data, and uses each group of sub-waterproof roll material image data as a continuous optimization processing object. The pixel values ​​in each optimization processing object are optimized, and then the pixel values ​​of the corresponding pixels in the atomic waterproof roll material image data are replaced with the optimized pixel values, until all pixels in all sub-waterproof roll material image data are replaced with the optimized pixel values, and then the sub-waterproof roll material image data are reorganized to obtain the optimized waterproof roll material image data.

[0020] Furthermore, when the segmentation module receives the waterproofing coiled material image data, a system end user manually selects a group of waterproofing coiled material image data that has been optimized by the optimization unit in the storage unit as a segmentation processing target;

[0021] The segmentation module is internally provided with submodules, including:

[0022] An analysis unit, used for traversing the waterproofing coiled material image data received by the segmentation module, and globally analyzing the complexity of the waterproofing coiled material image data based on the waterproofing coiled material image data;

[0023] The complexity analysis logic of the waterproofing coiled material image data in the analysis unit is expressed as follows:

[0024]

[0025] Where: θ is the complexity of the waterproofing membrane image data; p(i) is the probability of gray level i appearing in the waterproofing membrane image data; N edge is the number of edge pixels of the image; N total is the total number of image pixels;

[0026] The greater the complexity θ of the waterproof coiled material image data, the more complex the waterproof coiled material image data is and the more information it contains. Conversely, the simpler the waterproof coiled material image data is and the less information it contains.

[0027] Furthermore, when the segmentation module performs segmentation processing on the waterproof coiled material image data, the segmentation logic is set based on the complexity of the waterproof coiled material image data:

[0028] The size and aspect ratio of the sub-waterproofing roll material image data obtained by processing the waterproofing roll material image data based on the segmentation module are equal, and the number of sub-waterproofing roll material image data segmented on the long side and the wide side of the waterproofing roll material image data is equal. The more complex the waterproofing roll material image data is, the more sub-waterproofing roll material image data are segmented.

[0029] Furthermore, when the conversion module converts the sub-waterproofing coiled material image data into a grayscale image, the system end user sets the adjustment weight of each pixel in the sub-waterproofing coiled material image data based on the three channels of R, G, and B, and processes each pixel in the sub-waterproofing coiled material image data based on the set adjustment weight to output a grayscale image of the sub-waterproofing coiled material image data:

[0030]

[0031] Where: F is the pixel grayscale; R, G, B are the three-channel values ​​of the original pixel; ω1, ω2, ω3 are the adjustment weights;

[0032] Among them, the sum of ω1, ω2, and ω3 is 1.

[0033] Furthermore, the comprehensive similarity between all grayscale images in the evaluation module is evaluated by the following formula:

[0034]

[0035] Where: SIMM is the comprehensive similarity between all grayscale images; n is the set of a part of grayscale images in all grayscale images; m is the set of another part of grayscale images in all grayscale images; sim(p i ,p j ) is the similarity between the i-th group of grayscale images and the j-th group of grayscale images;

[0036] in, Table pair , n∪m represents all grayscale images, and the value of |nm| is 0 or 1.

[0037] Furthermore,

[0038] Where: M, N are the length and width of the grayscale image; is the gray value of the gray image pi at the position (x, y); is the grayscale value of the grayscale image pj at the (x, y) position.

[0039] Furthermore, a qualified determination threshold is set in the determination module, and the determination module runs based on the set qualified determination threshold and compares it with the SIMM to determine whether the waterproof coiled material image data source is qualified.

[0040] Furthermore, the control terminal is interactively connected to a collection module via a wireless network, the collection module is interactively connected to a storage unit and an optimization unit via a wireless network, the collection module is interactively connected to a segmentation module via a wireless network, the segmentation module is interactively connected to an analysis unit via a wireless network, the segmentation module and the optimization unit are interactively connected to the storage unit via a wireless network, and the segmentation module is interactively connected to a conversion module, an evaluation module and a determination module via a wireless network.

[0041] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0042] The present invention provides a waterproof membrane performance evaluation system based on machine vision. During operation, the system determines the performance of the waterproof membrane through image collection, analysis and evaluation of the waterproof membrane, thereby replacing manual inspection to a large extent, improving the production efficiency and pass rate of the waterproof membrane, and at the same time reducing the labor cost of the waterproof membrane inspection stage during the waterproof membrane production process, further improving the intelligence of the waterproof membrane production stage in the waterproof membrane industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a structural schematic diagram of a waterproof membrane performance evaluation system based on machine vision. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] The present invention will be further described below in conjunction with the embodiments.

[0047] Embodiment 1:

[0048] A waterproofing membrane performance evaluation system based on machine vision in this embodiment, such as Figure 1 As shown, including:

[0049] The control terminal is the main control terminal of the system and is used to issue execution commands;

[0050] An acquisition module is used to acquire image data of waterproof membranes;

[0051] The acquisition module is equipped with submodules, including:

[0052] A storage unit, used to receive the waterproofing coiled material image data collected by the collection module and store the waterproofing coiled material image data;

[0053] An optimization unit, used for selecting the waterproofing coiled material image data stored in the storage unit and optimizing the image data;

[0054] The acquisition module is integrated with an industrial camera with a resolution of 50 million pixels. When selecting the waterproof coiled material image data stored in the storage unit, the optimization unit selects in sequence based on the storage time sequence of the waterproof coiled material image data. Each time, a group of waterproof coiled material image data is selected, and after the selected waterproof coiled material image data is optimized, it is synchronously iterated to the storage position of the corresponding original waterproof coiled material image data in the storage unit.

[0055] The optimization processing logic for waterproofing membrane image data in the optimization unit is expressed as:

[0056]

[0057] Where: P(x, y) is the optimized image pixel value; P0(x, y) is the original image pixel value; k is the adjustment parameter; G x (x, y) is the horizontal gradient of the image; G y (x,y) is the vertical gradient of the image;

[0058] When the optimization unit optimizes the waterproofing coiled material image data, the waterproofing coiled material image data is first divided into equal parts to obtain a plurality of groups of sub-waterproofing coiled material image data, and each group of sub-waterproofing coiled material image data is used as a continuous optimization processing object. The pixel values ​​of the pixels in each optimization processing object are optimized, and then the pixel values ​​of the corresponding pixels in the atomic waterproofing coiled material image data are replaced with the optimized pixel values, until all the pixels in all the sub-waterproofing coiled material image data are replaced with the optimized pixel values, and then the sub-waterproofing coiled material image data are reorganized to obtain the waterproofing coiled material image data after optimization processing;

[0059] A segmentation module is used to receive the waterproofing coiled material image data collected by the collection module and perform segmentation processing on the waterproofing coiled material image data;

[0060] When the segmentation module receives the waterproofing coiled material image data, a system end user manually selects a group of waterproofing coiled material image data that has been optimized by the optimization unit in the storage unit as a segmentation processing target;

[0061] The segmentation module is internally provided with submodules, including:

[0062] An analysis unit, used for traversing the waterproofing coiled material image data received by the segmentation module, and globally analyzing the complexity of the waterproofing coiled material image data based on the waterproofing coiled material image data;

[0063] The complexity analysis logic of the waterproofing membrane image data in the analysis unit is expressed as:

[0064]

[0065] Where: θ is the complexity of the waterproofing membrane image data; p(i) is the probability of gray level i appearing in the waterproofing membrane image data; N edge is the number of edge pixels of the image; N total is the total number of image pixels;

[0066] The greater the complexity θ of the waterproofing coiled material image data, the more complex the waterproofing coiled material image data is and the more information it contains. Conversely, the simpler the waterproofing coiled material image data is and the less information it contains.

[0067] A conversion module, used for continuously receiving the sub-waterproofing coiled material image data obtained by segmentation in the segmentation module, and converting the sub-waterproofing coiled material image data into a grayscale image;

[0068] When the conversion module converts the sub-waterproofing membrane image data into a grayscale image, the system end user sets the adjustment weight of each pixel in the sub-waterproofing membrane image data based on the three channels of R, G, and B, and processes each pixel in the sub-waterproofing membrane image data based on the set adjustment weight to output a grayscale image of the sub-waterproofing membrane image data:

[0069]

[0070] Where: F is the pixel grayscale; R, G, B are the three-channel values ​​of the original pixel; ω1, ω2, ω3 are the adjustment weights;

[0071] Among them, the sum of ω1, ω2, and ω3 is 1;

[0072] An evaluation module is used to obtain all grayscale images converted by the conversion module and evaluate the comprehensive similarity between all grayscale images;

[0073] The comprehensive similarity between all grayscale images in the evaluation module is evaluated by the following formula:

[0074]

[0075] Where: SIMM is the comprehensive similarity between all grayscale images; n is the set of a part of grayscale images in all grayscale images; m is the set of another part of grayscale images in all grayscale images; sim(p i ,p j ) is the similarity between the i-th group of grayscale images and the j-th group of grayscale images;

[0076] in, Table pair , n∪m represents all grayscale images, and the value of |nm| is 0 or 1;

[0077] A determination module is used to receive the comprehensive similarity between all grayscale images, and determine whether the waterproof coiled material image data from the waterproof coiled material image data source is qualified based on the comprehensive similarity;

[0078] The control terminal is interactively connected to the acquisition module through a wireless network, the acquisition module is interactively connected to the storage unit and the optimization unit through a wireless network, the acquisition module is interactively connected to the segmentation module through a wireless network, the segmentation module is interactively connected to the analysis unit through a wireless network, the segmentation module and the optimization unit are interactively connected to the storage unit through a wireless network, and the segmentation module is interactively connected to the conversion module, the evaluation module and the judgment module through a wireless network.

[0079] In this embodiment, the control terminal controls the acquisition module to run and collect the image data of the waterproof coiled material, the storage unit synchronously receives the image data of the waterproof coiled material collected by the acquisition module, and stores the image data of the waterproof coiled material. The optimization unit selects the image data of the waterproof coiled material stored in the storage unit in real time and optimizes the image data. The segmentation module is post-operated to receive the image data of the waterproof coiled material collected by the acquisition module and segment the image data of the waterproof coiled material. The analysis unit synchronously traverses the image data of the waterproof coiled material received by the segmentation module, and globally analyzes the complexity of the image data of the waterproof coiled material based on the image data of the waterproof coiled material. The conversion module further continuously receives the sub-waterproof coiled material image data obtained by the segmentation module, and converts the sub-waterproof coiled material image data into a grayscale image. The evaluation module then obtains all the grayscale images converted by the conversion module, and evaluates the comprehensive similarity between all the grayscale images. Finally, the judgment module receives the comprehensive similarity between all the grayscale images, and judges whether the source of the waterproof coiled material image data is qualified based on the comprehensive similarity.

[0080] Through the operation of the system in the above embodiment, machine vision and waterproof membrane performance detection are combined to provide an effective, reliable and refined performance detection system for the waterproof membrane.

[0081] Embodiment 2:

[0082] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The waterproof membrane performance evaluation system based on machine vision in Example 1 is further described in detail:

[0083] When the segmentation module performs segmentation processing on the waterproofing membrane image data, the segmentation logic is set based on the complexity of the waterproofing membrane image data:

[0084] The size and aspect ratio of the sub-waterproofing roll material image data obtained by processing the waterproofing roll material image data based on the segmentation module are equal, and the number of sub-waterproofing roll material image data segmented on the long side and the wide side of the waterproofing roll material image data is equal. The more complex the waterproofing roll material image data is, the more sub-waterproofing roll material image data are segmented.

[0085]

[0086] Where: M, N are the length and width of the grayscale image; is the gray value of the gray image pi at the position (x, y); is the gray value of the gray image pj at the position (x, y);

[0087] A qualified determination threshold is set in the determination module. The determination module runs based on the set qualified determination threshold and compares it with the SIMM to determine whether the waterproofing membrane image data source is qualified.

[0088] In this embodiment, the above settings provide further operation logic support for the system in Embodiment 1, and provide a limited calculation formula for obtaining the similarity between two groups of grayscale images.

[0089] In summary, during the operation of the system in the above embodiment, the performance of the waterproof membrane is determined through image collection, analysis and evaluation of the waterproof membrane, which replaces manual inspection to a large extent, improves the production efficiency and pass rate of the waterproof membrane, and at the same time reduces the labor cost of the waterproof membrane inspection stage in the waterproof membrane production process, further improving the intelligence of the waterproof membrane production stage in the waterproof membrane industry.

[0090] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A waterproofing membrane performance evaluation system based on machine vision, characterized in that: include: The control terminal is the main control terminal of the system and is used to issue execution commands; An acquisition module is used to acquire image data of waterproof membranes; A segmentation module is used to receive the waterproofing coiled material image data collected by the collection module and perform segmentation processing on the waterproofing coiled material image data; A conversion module, used for continuously receiving the sub-waterproofing coiled material image data obtained by segmentation in the segmentation module, and converting the sub-waterproofing coiled material image data into a grayscale image; An evaluation module is used to obtain all grayscale images converted by the conversion module and evaluate the comprehensive similarity between all grayscale images; The determination module is used to receive the comprehensive similarity between all grayscale images, and determine whether the waterproof coiled material image data source is qualified based on the comprehensive similarity.

2. According to claim 1, a waterproofing membrane performance evaluation system based on machine vision is characterized in that: The acquisition module is provided with submodules at the lower level, including: A storage unit, used to receive the waterproofing coiled material image data collected by the collection module and store the waterproofing coiled material image data; An optimization unit, used for selecting the waterproofing coiled material image data stored in the storage unit and optimizing the image data; Among them, the acquisition module is integrated by an industrial camera with a resolution of 50 million pixels. When the optimization unit selects the waterproof coil material image data stored in the storage unit, it selects in sequence based on the storage sequence of the waterproof coil material image data, selects a group of waterproof coil material image data each time, and after the selected waterproof coil material image data is optimized, it synchronously iterates to the corresponding original waterproof coil material image data storage position in the storage unit.

3. The waterproofing membrane performance evaluation system based on machine vision according to claim 1 is characterized in that: The optimization processing logic for the waterproofing coiled material image data in the optimization unit is expressed as follows: Where: P(x, y) is the optimized image pixel value; P0(x, y) is the original image pixel value; k is the adjustment parameter; G x (x, y) is the horizontal gradient of the image; G y (x, y) is the vertical gradient of the image; Among them, when the optimization unit optimizes the waterproof roll material image data, it first divides the waterproof roll material image data into equal parts to obtain several groups of sub-waterproof roll material image data, and uses each group of sub-waterproof roll material image data as a continuous optimization processing object. The pixel values ​​in each optimization processing object are optimized, and then the pixel values ​​of the corresponding pixels in the atomic waterproof roll material image data are replaced with the optimized pixel values, until all pixels in all sub-waterproof roll material image data are replaced with the optimized pixel values, and then the sub-waterproof roll material image data are reorganized to obtain the optimized waterproof roll material image data.

4. The waterproofing membrane performance evaluation system based on machine vision according to claim 1 is characterized in that: When the segmentation module receives the waterproofing coiled material image data, a system end user manually selects a group of waterproofing coiled material image data optimized by the optimization unit in the storage unit as a segmentation processing target; The segmentation module is internally provided with submodules, including: An analysis unit, used for traversing the waterproofing coiled material image data received by the segmentation module, and globally analyzing the complexity of the waterproofing coiled material image data based on the waterproofing coiled material image data; The complexity analysis logic of the waterproofing coiled material image data in the analysis unit is expressed as follows: Where: θ is the complexity of the waterproofing membrane image data; p(i) is the probability of gray level i appearing in the waterproofing membrane image data; N edge is the number of edge pixels of the image; N total is the total number of image pixels; The greater the complexity θ of the waterproof coiled material image data, the more complex the waterproof coiled material image data is and the more information it contains. Conversely, the simpler the waterproof coiled material image data is and the less information it contains.

5. A waterproofing membrane performance evaluation system based on machine vision according to claim 1 or 4, characterized in that: When the segmentation module performs segmentation processing on the waterproof coiled material image data, the segmentation logic is set based on the complexity of the waterproof coiled material image data: The size and aspect ratio of the sub-waterproofing roll material image data obtained by processing the waterproofing roll material image data based on the segmentation module are equal, and the number of sub-waterproofing roll material image data segmented on the long side and the wide side of the waterproofing roll material image data is equal. The more complex the waterproofing roll material image data is, the more sub-waterproofing roll material image data are segmented.

6. The waterproofing membrane performance evaluation system based on machine vision according to claim 1 is characterized in that: When the conversion module converts the sub-waterproofing coiled material image data into a grayscale image, the system end user sets the adjustment weight of each pixel in the sub-waterproofing coiled material image data based on the three channels of R, G, and B, and processes each pixel in the sub-waterproofing coiled material image data based on the set adjustment weight to output a grayscale image of the sub-waterproofing coiled material image data: Where: F is the pixel grayscale; R, G, B are the three-channel values ​​of the original pixel; ω1, ω2, ω3 are the adjustment weights; Among them, the sum of ω1, ω2, and ω3 is 1.

7. The waterproofing membrane performance evaluation system based on machine vision according to claim 1 is characterized in that: The comprehensive similarity between all grayscale images in the evaluation module is evaluated by the following formula: Where: SIMM is the comprehensive similarity between all grayscale images; n is the set of a part of grayscale images in all grayscale images; m is the set of another part of grayscale images in all grayscale images; sim(p i ,p j ) is the similarity between the i-th group of grayscale images and the j-th group of grayscale images; in, Table pair , n∪m represents all grayscale images, and the value of |nm| is 0 or 1.

8. The waterproofing membrane performance evaluation system based on machine vision according to claim 7 is characterized in that: Where: M, N are the length and width of the grayscale image; is the gray value of the gray image pi at the position (x, y); is the grayscale value of the grayscale image pj at the (x, y) position.

9. The waterproofing membrane performance evaluation system based on machine vision according to claim 1, characterized in that: A qualified determination threshold is set in the determination module, and the determination module runs based on the set qualified determination threshold and compares it with the SIMM to determine whether the waterproof coiled material image data source is qualified.

10. The waterproofing membrane performance evaluation system based on machine vision according to claim 1, characterized in that: The control terminal is interactively connected to a collection module via a wireless network, the collection module is interactively connected to a storage unit and an optimization unit via a wireless network, the collection module is interactively connected to a segmentation module via a wireless network, the segmentation module is interactively connected to an analysis unit via a wireless network, the segmentation module and the optimization unit are interactively connected to the storage unit via a wireless network, and the segmentation module is interactively connected to a conversion module, an evaluation module and a determination module via a wireless network.

Images