Nondestructive testing method and system for tempered glass based on data analysis
Through data analysis and Hough circular transformation and other technologies, the gradient direction and sharpness of each pixel point in tempered glass are calculated, and thermal cracks and scratches are distinguished, which solves the problem of inaccurate detection in the existing technology and achieves more accurate tempering Glass inspection.
Patent Information
- Application Number
- CN202410940383.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-15
AI Technical Summary
The existing tempered glass detection methods are difficult to accurately distinguish between thermal cracks, scratches and spots, resulting in inaccurate detection results.
Using a data analysis method, the grayscale map of glass is obtained through Hoff circular transformation, the gradient direction and sharpness of each pixel point are calculated, and the characteristics and performance of thermal cracks are combined to distinguish thermal cracks and scratches to achieve accurate detection of glass. .
Effectively distinguish between thermal cracks and scratches, improve the accuracy of tempered glass detection and ensure the reliability of the detection results.
Smart Images

Figure CN118470025B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image data processing, and in particular to a method and system for nondestructive testing of tempered glass based on data analysis. Background Art
[0002] Tempered glass is a kind of glass that has been specially treated to have strong strength and impact resistance. In the production process of tempered glass, if the heating or cooling process is uneven, it may cause inconsistent temperature changes in different parts of the glass, resulting in thermal stress differences, causing thermal cracks inside the tempered glass. Thermal cracks in tempered glass reduce the strength and durability of the glass, making the glass more likely to break when subjected to pressure or impact, so it is necessary to detect thermal cracks in tempered glass in a timely manner.
[0003] In addition to thermal cracks, tempered glass may also have scratches and light spots. The grayscale values of pixels in the thermal crack area, scratch area, and light spot area are different from the grayscale values of pixels in the normal area. When the thermal cracks in the tempered glass are obtained through the edge detection algorithm, scratches and light spots may be obtained at the same time, which leads to inaccurate detection results of the tempered glass. Summary of the invention
[0004] The present invention provides a tempered glass nondestructive testing method and system based on data analysis to solve the existing problems.
[0005] The data analysis-based nondestructive testing method and system for tempered glass of the present invention adopts the following technical solutions:
[0006] The present invention proposes a nondestructive testing method for tempered glass based on data analysis, which comprises the following steps:
[0007] Acquire an image of the tempered glass to be inspected and obtain a glass grayscale image;
[0008] A circle in the glass grayscale image is obtained by Hough circle transform; several edges in the glass grayscale image are obtained by the grayscale value of each pixel in the glass grayscale image; several suspected thermal crack edges are obtained from the several edges according to the position of the circle in the glass grayscale image;
[0009] According to the grayscale values of each pixel point and the surrounding pixels in the glass grayscale image, the gradient direction of each pixel point is obtained; according to the gradient direction of each pixel point on each suspected thermal crack edge, the sharpness of each pixel point on each suspected thermal crack edge is obtained;
[0010] According to each suspected thermal crack edge, each suspected thermal crack region and the skeleton of each suspected thermal crack region are obtained; according to the gray value of each pixel point on the skeleton of each suspected thermal crack region and the difference in gray values of adjacent pixels on each suspected thermal crack edge, the thermal crack feature of each pixel point on each suspected thermal crack edge is obtained;
[0011] According to the change of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, as well as the sharpness and thermal crack characteristics of each pixel point on the edge of each suspected thermal crack, the thermal crack manifestation degree of each suspected thermal crack area is obtained, and the quality inspection of tempered glass is performed based on the thermal crack manifestation degree.
[0012] Furthermore, the step of acquiring the image of the tempered glass to be detected and obtaining the glass grayscale image includes the following specific steps:
[0013] An image of the tempered glass to be detected is obtained, and grayscale processing is performed on the image of the tempered glass to be detected to obtain a grayscale image of the tempered glass to be detected, which is recorded as a glass grayscale image.
[0014] Furthermore, the method of obtaining a plurality of suspected thermal crack edges from a plurality of edges according to the position of the circle in the glass grayscale image includes the following specific steps:
[0015] According to the position of the circle in the glass grayscale image, the edge of the circle in the glass grayscale image is obtained;
[0016] The edges of the circles in several edges of the glass grayscale image are removed to obtain several suspected thermal crack edges.
[0017] Furthermore, the step of obtaining the sharpness of each pixel point on each suspected thermal crack edge according to the gradient direction of each pixel point on each suspected thermal crack edge includes the following specific steps:
[0018] A tangent is drawn for each pixel on each suspected thermal crack edge, and the minimum angle between the gradient direction of any pixel on any suspected thermal crack edge and the tangent of the pixel is obtained, which is recorded as the minimum angle between the gradient direction of the pixel and the tangent;
[0019] According to the minimum angle between the gradient direction of each pixel point on each suspected thermal crack edge and the tangent line, the sharpness of each pixel point on each suspected thermal crack edge is obtained.
[0020] Furthermore, the step of obtaining the sharpness of each pixel point on each suspected thermal crack edge includes the following specific steps:
[0021]
[0022] In the formula, Indicates The first The sharpness of each pixel, Indicates 180 degrees, Indicates The first The minimum angle between the gradient direction of a pixel and the tangent line, is a logarithmic function with base 2.
[0023] Furthermore, the thermal crack feature of each pixel on each suspected thermal crack edge is obtained according to the gray value of each pixel on the skeleton of each suspected thermal crack area and the difference in gray values of adjacent pixels on each suspected thermal crack edge, and the specific steps include the following:
[0024] The first The area within the edge of the suspected thermal crack is recorded as suspected thermal crack areas;
[0025] Get the The skeleton of the suspected thermal crack area; The skeleton of the suspected hot crack area is The first The pixel with the shortest Euclidean distance is denoted as The first The corresponding points of pixels;
[0026] The thermal crack feature of each pixel on each suspected thermal crack edge is obtained according to the gray value of the corresponding point of each pixel on each suspected thermal crack edge and the difference in gray value between each pixel on each suspected thermal crack edge and the surrounding pixels.
[0027] Furthermore, the step of obtaining the thermal crack feature of each pixel point on each suspected thermal crack edge includes the following specific steps:
[0028]
[0029] In the formula, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, represents the absolute value function, Indicates The first The gray value of the corresponding point of each pixel.
[0030] Furthermore, the degree of thermal crack performance of each suspected thermal crack region is obtained according to the change in the gradient direction of the pixel points on the skeleton of each suspected thermal crack region, and the sharpness and thermal crack characteristics of each pixel point on the edge of each suspected thermal crack, including the following specific steps:
[0031] According to the gradient direction of each pixel point on the skeleton of each suspected thermal crack region, the variance of the gradient direction of the pixel point on the skeleton of each suspected thermal crack region is obtained;
[0032] According to the sharpness of each edge pixel point on each suspected thermal crack edge, the thermal crack characteristics and the variance of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, the thermal crack performance degree of each suspected thermal crack area is obtained. The specific calculation formula is as follows:
[0033]
[0034] In the formula, Indicates The degree of thermal cracking in the suspected thermal crack area, Indicates The number of pixels on the edge of a suspected thermal crack, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The sharpness of each pixel, Indicates The variance of the gradient direction of the pixel points on the skeleton of the suspected thermal crack area, Represents the sigmoid function.
[0035] Furthermore, the quality inspection of the tempered glass based on the degree of thermal cracking includes the following specific steps:
[0036] Preset performance threshold , the suspected thermal crack area with a thermal crack manifestation degree greater than or equal to the manifestation degree threshold is recorded as the thermal crack area;
[0037] Preset area ratio threshold , if the ratio of the sum of the number of pixels in all thermal crack regions to the total number of pixels in the glass grayscale image is greater than or equal to , the quality of the tempered glass to be tested is unqualified.
[0038] The present invention also proposes a tempered glass nondestructive testing system based on data analysis, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned tempered glass nondestructive testing method based on data analysis.
[0039] The technical solution of the present invention has the following beneficial effects: when obtaining a thermal crack region from a glass grayscale image, firstly, the edge in the glass grayscale image is obtained by a sobel operator; based on the fact that the shape of the light spot in the glass grayscale image is close to a circle, the edge of the light spot is distinguished from the edge of the thermal crack; based on the fact that the scratch edge is relatively smooth and the thermal crack edge is relatively sharp, the sharpness of the pixel points on the thermal crack edge is much greater than the sharpness of the pixel points on the scratch edge; based on the fact that the thermal crack is inside the glass and the scratch is on the glass surface, combined with the fact that the thermal crack has branches, when obtaining the glass grayscale image, due to the reflection and diffusion of light by the branches, the grayscale values of the thermal crack edge pixel points and the surrounding edge pixel points have a large difference, so that the thermal crack characteristics of the thermal crack edge pixel points are much greater than the thermal crack characteristics of the scratch edge pixel points; based on the fact that the thermal crack is relatively curved and the scratch is relatively straight, combined with the thermal crack characteristics and sharpness of each pixel point on each suspected thermal crack edge, the thermal crack region is distinguished from the scratch region, so that the obtained thermal crack region is more accurate, resulting in a more accurate detection result of the tempered glass. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] 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 description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0041] Figure 1 It is a flowchart of the steps of the nondestructive testing method of tempered glass based on data analysis of the present invention;
[0042] Figure 2 For thermal cracks and scratches images. DETAILED DESCRIPTION
[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the method and system for nondestructive testing of tempered glass based on data analysis proposed by the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0045] The specific scheme of the method and system for nondestructive testing of tempered glass based on data analysis provided by the present invention is described in detail below with reference to the accompanying drawings.
[0046] See also Figure 1 , which shows a flowchart of a method for nondestructive testing of tempered glass based on data analysis provided by an embodiment of the present invention, the method comprising the following steps:
[0047] Step S001: Acquire an image of the tempered glass to be inspected and obtain a glass grayscale image.
[0048] The main purpose of this embodiment is to obtain thermal cracks in the tempered glass, and thus obtain an image of the tempered glass.
[0049] Specifically, an industrial camera is used to obtain an image of the tempered glass to be detected, and the image of the tempered glass to be detected is gray-scaled to obtain a grayscale image of the tempered glass to be detected, which is recorded as a glass grayscale image.
[0050] Step S002: Obtain a circle in the glass grayscale image through Hough circle transform; obtain a number of edges in the glass grayscale image through the grayscale value of each pixel in the glass grayscale image; obtain a number of suspected thermal crack edges from the several edges according to the position of the circle in the glass grayscale image.
[0051] It should be noted that the grayscale values of the thermal crack pixels, the light spot pixels, and the scratch pixels in the glass grayscale image are different from the grayscale values of normal pixels, so when the edge detection algorithm is used to obtain the thermal crack edge in the glass grayscale image, the thermal crack edge, the scratch edge, and the light spot edge may be obtained at the same time. Therefore, the edge detection results should be analyzed to obtain the thermal crack area in the glass grayscale image.
[0052] It should be further explained that the shape of the light spot in the tempered glass is close to a circle, while the shape of the thermal crack in the tempered glass is quite different from a circle. Therefore, according to the position of the circle in the glass grayscale image, the edge of the suspected thermal crack in the glass grayscale image is obtained.
[0053] Specifically, several edges in the glass grayscale image are obtained by using an edge detection algorithm, wherein the edge detection algorithm is a well-known technology and will not be described in detail in this embodiment.
[0054] Further, the glass grayscale image is subjected to Hough circle transformation to obtain the circles in the glass grayscale image. The Hough circle transformation is a well-known technology and will not be described in detail in this embodiment. According to the position of the circle in the glass grayscale image, the edge of the circle in the glass grayscale image is obtained; the edge of the circle in several edges in the glass grayscale image is removed to obtain several suspected thermal crack edges.
[0055] At this point, several suspected thermal crack edges are obtained.
[0056] Step S003: Obtain the gradient direction of each pixel point according to the grayscale values of each pixel point and the surrounding pixels in the glass grayscale image; obtain the sharpness of each pixel point on each suspected thermal crack edge according to the gradient direction of each pixel point on each suspected thermal crack edge.
[0057] It should be noted that the thermal cracks in the glass extend along the path of grain boundaries or internal defects, making the edges of the thermal cracks sharper. The scratches on the glass are caused by the relative sliding of the glass and the heavy object, making the edges of the scratches smoother. Therefore, the sharpness of each pixel on the edge of each suspected thermal crack is calculated.
[0058] It should be further explained that the horizontal gradient of a pixel reflects the grayscale difference between the pixel and the pixel in its horizontal direction, and the vertical gradient of a pixel reflects the grayscale difference between the pixel and the pixel in its vertical direction. When an area has both sharp edges and smooth edges, the distribution of grayscale values of the pixels around the sharp edge is different from the distribution of grayscale values of the pixels around the smooth edge, so that the horizontal gradient value and vertical gradient value of the pixel on the sharp edge are different from the horizontal gradient value and vertical gradient value of the pixel on the smooth edge. And the gradient direction of the pixel is obtained by the horizontal gradient and vertical gradient of the pixel. Therefore, according to the gradient direction of each pixel on each suspected thermal crack edge, the sharpness of each pixel on each suspected thermal crack edge is obtained. Among them, the image of thermal cracks and scratches in glass is as follows. Figure 2 shown.
[0059] Specifically, the gradient direction of each pixel in the glass grayscale image is obtained by using the Sobel operator. The Sobel operator is a well-known technique and is not described in detail in this embodiment. A tangent is made to each pixel on each suspected thermal crack edge to obtain the minimum angle between the gradient direction of any pixel on any suspected thermal crack edge and the tangent of the pixel, which is recorded as the minimum angle between the gradient direction of the pixel and the tangent. Obtaining the tangent of each pixel is a well-known technique and is not described in detail in this embodiment.
[0060] Furthermore, according to the minimum angle between the gradient direction of each pixel point on each suspected thermal crack edge and the tangent line of the pixel point, the sharpness of each pixel point on each suspected thermal crack edge is obtained. The specific calculation formula is as follows:
[0061]
[0062] In the formula, Indicates The first The sharpness of each pixel, Indicates 180 degrees, Indicates The first The minimum angle between the gradient direction of a pixel and the tangent line, is a logarithmic function with base 2, is the input of the model.
[0063] It should be noted that The larger the value, the The first The sharper the pixel, the The more likely a suspected thermal crack edge is to be a thermal crack edge.
[0064] At this point, the sharpness of each pixel point on each suspected thermal crack edge is obtained.
[0065] Step S004: Based on each suspected thermal crack edge, obtain each suspected thermal crack area and the skeleton of each suspected thermal crack area; based on the grayscale value of each pixel point on the skeleton of each suspected thermal crack area and the difference in grayscale values of adjacent pixels on each suspected thermal crack edge, obtain the thermal crack feature of each pixel point on each suspected thermal crack edge.
[0066] It should be noted that the scratches on the glass are caused by the relative sliding of the glass with other heavy objects, so the scratches are mostly strip-shaped. The depth of the scratch is mainly determined by the friction force, which is mainly determined by the mass of the heavy object, the tilt angle of the heavy object and the friction coefficient. In addition, during the relative sliding process, the mass of the heavy object, the tilt angle of the heavy object and the friction coefficient remain almost unchanged, making the grayscale values of each pixel on the edge of a scratch in the glass grayscale image relatively close.
[0067] It should be further explained that the thermal cracks are inside the glass, so when the glass image is acquired, the grayscale values of the surrounding pixels will be affected by the thermal cracks due to reasons such as light scattering and diffusion. Thermal cracks are formed due to thermal stress caused by temperature changes in the glass, so the location of the thermal cracks in the glass is relatively random, but the depths of the edges at different locations on the same thermal crack are relatively similar. In addition, the thermal cracks have branches, and the shortest Euclidean distances from different edge pixels in each branch of the thermal crack to the edge pixels of another branch are different, so that when the image of the tempered glass is acquired, each branch has a different effect on the grayscale values of different edge pixels in other branches, resulting in a large difference in the grayscale values of the pixels on the edges of the thermal crack branches. Therefore, based on the difference in the grayscale value of each pixel on each suspected thermal crack edge and the grayscale value of the surrounding pixels, the thermal crack characteristics of each pixel on each suspected thermal crack edge are obtained.
[0068] It should be noted that the thermal crack is inside the glass, while the scratch is on the surface of the glass, so the grayscale value of the thermal crack pixel in the glass grayscale image is smaller than the grayscale value of the scratch pixel. Therefore, the thermal crack feature of each pixel on each suspected thermal crack edge is calculated based on the grayscale value of each pixel in the area within the suspected thermal crack edge.
[0069] Specifically, the area within each suspected thermal crack edge in the glass grayscale image is recorded as the suspected thermal crack area, where The area within the edge of the suspected thermal crack is Obtain the skeleton of each suspected thermal crack region. Obtaining the skeleton of each region is a well-known technique in morphological algorithms and will not be described in detail in this embodiment. The skeleton of the suspected hot crack area is The first The pixel with the shortest Euclidean distance is denoted as The first The corresponding points of pixels.
[0070] Furthermore, according to the gray value of the corresponding point of each pixel on each suspected thermal crack edge and the difference in gray value between each pixel on each suspected thermal crack edge and the surrounding pixels, the thermal crack feature of each pixel on each suspected thermal crack edge is obtained. The specific calculation formula is as follows:
[0071]
[0072] In the formula, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, represents the absolute value function, Indicates The first The gray value of the corresponding point of each pixel.
[0073] It should be noted that The smaller the value, the The first The greater the probability that the pixel is inside the glass, the The first The larger the thermal crack characteristics of each pixel; The larger the value is, the more The first The greater the possibility that the pixel is affected by the thermal crack branch, the greater the probability that the The first The larger the thermal crack characteristics of each pixel.
[0074] At this point, the thermal crack feature of each pixel point on each suspected thermal crack edge is obtained.
[0075] Step S005: According to the change in the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, as well as the sharpness and thermal crack characteristics of each pixel point on the edge of each suspected thermal crack, the thermal crack manifestation degree of each suspected thermal crack area is obtained, and the quality inspection of the tempered glass is performed based on the thermal crack manifestation degree.
[0076] It should be noted that the thermal cracks in the glass are formed inside the glass, and the thermal cracks extend along the path of grain boundaries or internal defects, which makes the thermal cracks in the glass have a certain curvature; the scratches on the glass are caused by the relative sliding of the glass and the heavy object, which makes the scratches in the glass close to a straight line; thus, the gradient direction of the pixel points on the skeleton of the thermal crack area in the glass grayscale image has a strong volatility, while the gradient direction of the pixel points on the skeleton of the scratch area is relatively similar. Therefore, according to the volatility of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, the thermal crack performance characteristics of each suspected thermal crack area are obtained.
[0077] According to the gradient direction of each pixel point on the skeleton of each suspected thermal crack region, the variance of the gradient direction of the pixel point on the skeleton of each suspected thermal crack region is obtained.
[0078] According to the sharpness of each edge pixel point on each suspected thermal crack edge, the thermal crack characteristics and the variance of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, the thermal crack performance degree of each suspected thermal crack area is obtained. The specific calculation formula is as follows:
[0079]
[0080] In the formula, Indicates The degree of thermal cracking in the suspected thermal crack area, Indicates The number of pixels on the edge of a suspected thermal crack, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The sharpness of each pixel, Indicates The variance of the gradient direction of the pixel points on the skeleton of the suspected thermal crack area, Represents the sigmoid function, which is used for normalization processing in this embodiment.
[0081] It should be noted that The larger the value, the The stronger the bending of the suspected thermal crack area, the The greater the degree of thermal cracking in the suspected thermal crack area; The larger the value, the The greater the thermal crack performance of the pixel point on the edge of the suspected thermal crack area, the greater the thermal crack performance of the pixel point on the edge of the suspected thermal crack area. The greater the degree of thermal crack manifestation in the suspected thermal crack area.
[0082] Furthermore, a preset performance threshold ,if , then the The suspected thermal crack area is recorded as the thermal crack area. , this is used as an example for description, and other values may be set in other implementation modes.
[0083] Furthermore, a preset area ratio threshold , if the ratio of the sum of the number of pixels in all thermal crack regions to the total number of pixels in the glass grayscale image is greater than or equal to , then the quality of the tempered glass to be tested is unqualified. The area ratio threshold preset in this embodiment is , this is used as an example for description, and other values may be set in other implementation modes.
[0084] At this point, this embodiment is completed.
[0085] The present invention also provides a tempered glass nondestructive testing system based on data analysis, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned tempered glass nondestructive testing method based on data analysis.
[0086] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A nondestructive testing method for tempered glass based on data analysis, characterized in that: The method comprises the following steps: Acquire an image of the tempered glass to be inspected and obtain a glass grayscale image; A circle in the glass grayscale image is obtained by Hough circle transform; several edges in the glass grayscale image are obtained by the grayscale value of each pixel in the glass grayscale image; several suspected thermal crack edges are obtained from the several edges according to the position of the circle in the glass grayscale image; According to the grayscale values of each pixel point and the surrounding pixels in the glass grayscale image, the gradient direction of each pixel point is obtained; according to the gradient direction of each pixel point on each suspected thermal crack edge, the sharpness of each pixel point on each suspected thermal crack edge is obtained; According to each suspected thermal crack edge, each suspected thermal crack region and the skeleton of each suspected thermal crack region are obtained; according to the gray value of each pixel point on the skeleton of each suspected thermal crack region and the difference in gray values of adjacent pixels on each suspected thermal crack edge, the thermal crack feature of each pixel point on each suspected thermal crack edge is obtained; According to the change of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, as well as the sharpness and thermal crack characteristics of each pixel point on the edge of each suspected thermal crack, the thermal crack manifestation degree of each suspected thermal crack area is obtained, and the quality inspection of the tempered glass is performed based on the thermal crack manifestation degree; The step of obtaining the sharpness of each pixel point on each suspected thermal crack edge according to the gradient direction of each pixel point on each suspected thermal crack edge includes the following specific steps: A tangent is drawn for each pixel on each suspected thermal crack edge, and the minimum angle between the gradient direction of any pixel on any suspected thermal crack edge and the tangent of the pixel is obtained, which is recorded as the minimum angle between the gradient direction of the pixel and the tangent; According to the minimum angle between the gradient direction of each pixel point on each suspected thermal crack edge and the tangent line, the sharpness of each pixel point on each suspected thermal crack edge is obtained; The specific steps of obtaining the sharpness of each pixel point on each suspected thermal crack edge are as follows: In the formula, Indicates The first The sharpness of each pixel, Indicates 180 degrees, Indicates The first The minimum angle between the gradient direction of a pixel and the tangent line, It is a logarithmic function with base 2; The method of obtaining the thermal crack feature of each pixel on each suspected thermal crack edge according to the gray value of each pixel on the skeleton of each suspected thermal crack region and the difference in gray values of adjacent pixels on each suspected thermal crack edge includes the following specific steps: The first The area within the edge of the suspected thermal crack is recorded as suspected thermal crack areas; Get the The skeleton of the suspected thermal crack area; The skeleton of the suspected hot crack area is The first The pixel with the shortest Euclidean distance is denoted as The first The corresponding points of pixels; According to the gray value of the corresponding point of each pixel point on each suspected thermal crack edge and the difference in gray value between each pixel point on each suspected thermal crack edge and the surrounding pixel points, the thermal crack feature of each pixel point on each suspected thermal crack edge is obtained; The specific steps of obtaining the thermal crack feature of each pixel point on each suspected thermal crack edge are as follows: In the formula, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, Indicates The first The gray value of a pixel, represents the absolute value function, Indicates The first The gray value of the corresponding point of each pixel; The specific steps of obtaining the degree of thermal crack performance of each suspected thermal crack region according to the change of the gradient direction of the pixel points on the skeleton of each suspected thermal crack region and the sharpness and thermal crack characteristics of each pixel point on the edge of each suspected thermal crack are as follows: According to the gradient direction of each pixel point on the skeleton of each suspected thermal crack region, the variance of the gradient direction of the pixel point on the skeleton of each suspected thermal crack region is obtained; According to the sharpness of each edge pixel point on each suspected thermal crack edge, the thermal crack characteristics and the variance of the gradient direction of the pixel points on the skeleton of each suspected thermal crack area, the thermal crack performance degree of each suspected thermal crack area is obtained. The specific calculation formula is as follows: In the formula, Indicates The degree of thermal cracking in the suspected thermal crack area, Indicates The number of pixels on the edge of a suspected thermal crack, Indicates The first The thermal crack characteristics of each pixel, Indicates The first The sharpness of each pixel, Indicates The variance of the gradient direction of the pixel points on the skeleton of the suspected thermal crack area, Represents the sigmoid function; The quality inspection of tempered glass based on the degree of thermal cracking includes the following specific steps: Preset performance threshold , the suspected thermal crack area with a thermal crack manifestation degree greater than or equal to the manifestation degree threshold is recorded as the thermal crack area; Preset area ratio threshold , if the ratio of the sum of the number of pixels in all thermal crack regions to the total number of pixels in the glass grayscale image is greater than or equal to , the quality of the tempered glass to be tested is unqualified.
2. The method for nondestructive testing of tempered glass based on data analysis according to claim 1, characterized in that: The step of acquiring the image of the tempered glass to be detected and obtaining the glass grayscale image includes the following specific steps: An image of the tempered glass to be detected is obtained, and grayscale processing is performed on the image of the tempered glass to be detected to obtain a grayscale image of the tempered glass to be detected, which is recorded as a glass grayscale image.
3. The method for nondestructive testing of tempered glass based on data analysis according to claim 1, characterized in that: The specific steps of obtaining a plurality of suspected thermal crack edges from a plurality of edges according to the position of the circle in the glass grayscale image are as follows: According to the position of the circle in the glass grayscale image, the edge of the circle in the glass grayscale image is obtained; The edges of the circles in several edges of the glass grayscale image are removed to obtain several suspected thermal crack edges.
4. A tempered glass nondestructive testing system based on data analysis, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method for nondestructive testing of tempered glass based on data analysis as described in any one of claims 1 to 3 are implemented.
