Glass detection method and system based on artificial intelligence

Abstract

The invention discloses a glass detection method and system based on artificial intelligence, and relates to the technical field of image analysis, and the method comprises the steps: based on an intelligent analysis terminal, controlling shooting equipment to carry out image collection on glass, and obtaining a to-be-detected glass image; analyzing and processing the to-be-detected glass image to obtain real size data of the to-be-detected glass; and performing feature analysis processing on the to-be-detected glass image, and determining position coordinates of defects of the to-be-detected glass. The method comprises the following steps: firstly, carrying out image analysis on a to-be-detected glass image to determine the position of a defect in the glass, then, calculating the central positions of the defect and the glass to determine the distance between the defect and the central position of the glass, and finally, judging the distance between the defect and the central position of the glass to determine the quality of the glass. And the glass can be cut according to the distance between the defect and the center position of the glass to remove the defect on the glass, so that the glass with the defect is prevented from being abandoned, and the production cost of the glass is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a glass detection method and system based on artificial intelligence. Background Art

[0002] Glass (foreign name: Glass) is an amorphous inorganic non-metallic material, which is generally made of a variety of inorganic minerals as the main raw materials, and a small amount of auxiliary raw materials. Its main components are silicon dioxide and other oxides. It can be divided into quartz glass, high-pressure resistant glass, UV-proof glass and explosion-proof glass, etc., and is widely used in construction, daily use, art, medical treatment, chemistry, electronics and instrumentation.

[0003] Defects may occur in glass during production. When used, the overall rigidity of defective glass may decrease. If the defective glass is discarded directly, the production cost of the glass will increase. Summary of the invention

[0004] In order to solve the above technical problems, a glass detection method and system based on artificial intelligence are provided. This technical solution solves the problem that defects may occur in the glass during production proposed in the above background technology. When the defective glass is used, the overall rigidity may decrease. If the defective glass is discarded directly, the production cost of the glass will increase.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: A glass detection method based on artificial intelligence, comprising: Based on the intelligent analysis terminal, control the shooting equipment to collect images of the glass and obtain the image of the glass to be tested; Based on the intelligent analysis terminal, the image of the glass to be inspected is analyzed and processed to obtain the real size data of the glass to be inspected; Based on the intelligent analysis terminal, the feature analysis and processing of the glass image to be inspected is performed to determine the position coordinates of the defects of the glass to be inspected; Based on the intelligent analysis terminal, the coordinates of the defect positions of the glass to be inspected are analyzed and processed to determine the quality of the glass to be inspected.

[0006] Preferably, the intelligent analysis terminal is used to control the shooting device to collect images of the glass, and obtaining the image of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the database system is processed for information extraction to obtain standard glass images; Based on the intelligent analysis terminal, image parameter analysis and processing are performed on the standard glass image to obtain the shooting parameters and shooting background of the standard glass image; The intelligent analysis terminal adjusts the parameters of the shooting device according to the shooting parameters of the standard glass image to obtain the calibration shooting device; The glass to be inspected is placed in the shooting background, and based on the intelligent analysis terminal, the correction shooting equipment is controlled to collect images of the glass to be inspected to obtain the image of the glass to be inspected.

[0007] Preferably, the step of analyzing and processing the image of the glass to be detected based on the intelligent analysis terminal to obtain the real size data of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the image of the glass to be inspected is measured and processed to obtain the scaled size data of the glass to be inspected; Based on the intelligent analysis terminal, the parameters of the calibration shooting equipment are extracted and processed to obtain the image scaling ratio; Based on the intelligent analysis terminal, the scaled size data of the glass to be inspected and the scaled ratio of the image are calculated and processed to obtain the real size data of the glass to be inspected; The specific calculation formula for obtaining the real size data of the glass to be tested is: ; In the formula, The actual size data of the glass to be tested; The scaled size data of the glass to be inspected; The image scaling factor.

[0008] Preferably, the method of performing feature analysis on the glass image to be detected based on the intelligent analysis terminal to determine the coordinates of the position of the defect in the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the image of the glass to be inspected is analyzed and processed to determine the center position of the glass to be inspected; Based on the intelligent analysis terminal, the center position of the glass to be detected is set as the coordinate origin, the horizontal direction of the center position of the glass to be detected is set as the X-axis, and the vertical direction of the center position of the glass is set as the Y-axis to construct a rectangular coordinate system; Based on the intelligent analysis terminal, the feature analysis and processing of the glass image to be inspected is performed to determine the location of the defect in the glass to be inspected; The intelligent analysis terminal places the position of the glass defect to be detected in a rectangular coordinate system to determine the coordinates of the position of the glass defect to be detected.

[0009] Preferably, the step of analyzing and processing the image of the glass to be detected based on the intelligent analysis terminal to determine the center position of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, feature extraction and processing are performed on the glass image to be inspected to determine the edge position of the glass to be inspected; The intelligent analysis terminal draws a straight line from the edge position of the glass to be detected to the edge position of other glass to be detected, and obtains several groups of straight line intersection points; Based on the intelligent analysis terminal, several groups of straight line intersections are screened and processed to determine the center position of the glass to be tested; The specific calculation formula for determining the center position of the glass to be detected is:

[0010] In the formula, is the center position of the glass to be tested; is the number of straight lines that each straight line intersection passes through; n is the specific number of several groups of straight line intersections; max() is the maximum value function.

[0011] Preferably, the method of performing feature analysis on the glass image to be detected based on the intelligent analysis terminal to determine the position of the defect of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the standard glass image and the glass image to be detected are binarized to obtain the standard glass grayscale image and the glass grayscale image to be detected; Based on the intelligent analysis terminal, grayscale values ​​of the standard glass grayscale image and the glass grayscale image to be detected are calculated respectively to obtain the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected; Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected are judged and processed to determine the position of the defect in the glass to be detected.

[0012] Preferably, the intelligent analysis terminal is used to judge the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected, and determine the position of the defect of the glass to be detected, which specifically includes the following steps: Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be tested are judged; If the grayscale value of each pixel in the standard glass grayscale image is consistent with the grayscale value of each pixel in the grayscale image of the glass to be tested, the glass to be tested has no defects; If the grayscale value of each pixel in the standard glass grayscale image is inconsistent with the grayscale value of each pixel in the glass grayscale image to be detected, the position where the grayscale value of the pixel is inconsistent is recorded, and the position where the grayscale value of the pixel is inconsistent is set as the position of the glass defect to be detected.

[0013] Preferably, the method of analyzing and processing the coordinates of the defect position of the glass to be detected based on the intelligent analysis terminal to determine the quality of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the coordinates of the position of the glass defect to be detected and the coordinate origin are calculated and processed to determine the Euclidean distance of the glass defect to be detected; Based on the intelligent analysis terminal, the database system is used to extract data and obtain glass quality evaluation standards; Based on the intelligent analysis terminal, the glass quality assessment standard and the actual size data of the glass to be tested are analyzed and processed to set the defect distance threshold; Based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are compared and judged to determine the quality of the glass to be detected; The specific calculation formula of the Euclidean distance of the glass defect to be detected is:

[0014] In the formula, is the Euclidean distance of the glass defect to be detected; x is the X-axis coordinate of the position of the glass defect to be detected; y is the Y-axis coordinate of the position of the glass defect to be detected.

[0015] Preferably, the method of comparing and judging the defect distance threshold and the Euclidean distance of the glass defect to be detected based on the intelligent analysis terminal to determine the quality of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are judged and processed; If the Euclidean distance of the glass defect to be detected is less than or equal to the first threshold of the defect distance, the quality of the glass to be detected is poor; If the Euclidean distance of the glass defect to be detected is greater than the first threshold of the defect distance, and the Euclidean distance of the glass defect to be detected is less than or equal to the second threshold of the defect distance, the quality of the glass to be detected is good; If the Euclidean distance of the defect in the glass to be detected is greater than the second threshold of the defect distance, the quality of the glass to be detected is excellent.

[0016] Furthermore, a glass detection system based on artificial intelligence is proposed, which is used to implement the above-mentioned glass detection method based on artificial intelligence, including: An intelligent analysis terminal, which is used to control data transmission and information interaction between various modules; A database system, wherein the database system is used to store standard glass images and glass quality assessment standards; A parameter extraction module, wherein the parameter extraction module is used to perform parameter analysis on the standard glass image to determine the shooting parameters and shooting background of the standard glass image; A shooting device, wherein the shooting device is used to collect images of the glass to be inspected and obtain images of the glass to be inspected; A size calculation module, which performs calculations based on the scaled size data of the glass to be detected and the scaled ratio of the image to obtain the real size data of the glass to be detected; A coordinate system building module, wherein the coordinate system building module builds a rectangular coordinate system according to the center position of the glass to be detected; An image analysis module, which performs feature analysis on the image of the glass to be inspected to determine whether the glass to be inspected has defects; A data calculation module, which calculates and processes the position coordinates of the glass defect to be detected and the coordinate origin to determine the Euclidean distance of the glass defect to be detected; The quality judgment module compares and judges the defect distance threshold and the Euclidean distance of the defect of the glass to be detected to determine the quality of the glass to be detected.

[0017] Compared with the prior art, the present invention provides a glass detection method and system based on artificial intelligence, which has the following beneficial effects: The present invention first performs image analysis on the glass image to be inspected to determine the position of the defect in the glass, then calculates the center position of the defect and the glass to determine the distance between the defect and the center position of the glass, and finally judges the distance between the defect and the center position of the glass to determine the quality of the glass. In addition, the glass can also be cut according to the distance between the defect and the center position of the glass to remove the defect on the glass. The above method avoids discarding glass with defects and reduces the production cost of the glass. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of steps S100-S400 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 2 This is a flow chart of steps S101-S104 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 3 This is a flow chart of steps S201-S203 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 4 This is a flow chart of steps S301-S304 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 5 This is a flow chart of steps S3011-S3013 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 6 This is a flow chart of steps S3031-S3033 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 7 This is a flow chart of steps S30331-S30333 in a glass detection method based on artificial intelligence proposed by the present invention; Figure 8 This is a flow chart of steps S401-S404 in a glass detection method based on artificial intelligence proposed by the present invention; Fig. 9 This is a flow chart of steps S4041-S4044 in a glass detection method based on artificial intelligence proposed by the present invention; Fig.10 This is a structural block diagram of an artificial intelligence-based glass detection system proposed by the present invention. DETAILED DESCRIPTION

[0019] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0020] Reference Figure 1 As shown, a glass detection method based on artificial intelligence includes: S100, based on the intelligent analysis terminal, controlling the shooting device to collect images of the glass to obtain images of the glass to be detected; S200, analyzing and processing the image of the glass to be inspected based on the intelligent analysis terminal to obtain the real size data of the glass to be inspected; S300, based on the intelligent analysis terminal, feature analysis and processing are performed on the glass image to be detected to determine the position coordinates of the defects of the glass to be detected; S400, based on the intelligent analysis terminal, analyzing and processing the coordinates of the defect position of the glass to be detected, and determining the quality of the glass to be detected; It is understandable to those skilled in the art that defects may occur in glass during its production, such as bubbles inside the glass or depressions on the glass surface. If the defective glass is directly discarded, the production cost of the glass will increase. Therefore, by judging the distance between the glass defect and the center position of the glass, a method for cutting the glass defect is designed so that the cut glass does not contain defects, thereby avoiding the discarding of defective glass and reducing the production cost of the glass.

[0021] Reference Figure 2 As shown, based on the intelligent analysis terminal, controlling the shooting device to collect images of the glass, and obtaining the image of the glass to be detected specifically includes the following steps: S101, extracting information from the database system based on the intelligent analysis terminal to obtain a standard glass image; S102, performing image parameter analysis and processing on the standard glass image based on the intelligent analysis terminal to obtain shooting parameters and shooting background of the standard glass image; S103, the intelligent analysis terminal adjusts the parameters of the shooting device according to the shooting parameters of the standard glass image to obtain a calibrated shooting device; S104, placing the glass to be inspected in the shooting background, and based on the intelligent analysis terminal, controlling the calibration shooting device to collect images of the glass to be inspected, and obtaining an image of the glass to be inspected; In this embodiment, when two images need to be compared, they need to have the same shooting background and the same shooting parameters. Therefore, data is read from the database system to obtain a standard glass image, and then parameters are extracted from the standard glass image to obtain the shooting parameters and shooting background of the standard glass image. Finally, the glass to be inspected is placed in the shooting background, and then the image of the glass to be inspected is captured by the shooting device with adjusted parameters. In this way, the two images with the same shooting background and shooting parameters can be compared, and the defects of the glass image to be inspected are indirectly determined through the standard glass image.

[0022] Reference Figure 3 As shown, based on the intelligent analysis terminal, analyzing and processing the image of the glass to be detected to obtain the real size data of the glass to be detected specifically includes the following steps: S201, based on the intelligent analysis terminal, measuring and processing the image of the glass to be inspected, and obtaining the scaled size data of the glass to be inspected; S202, based on the intelligent analysis terminal, performing parameter extraction processing on the calibration shooting device to obtain the zoom ratio of the image; S203, based on the intelligent analysis terminal, calculating and processing the scaled size data of the glass to be inspected and the scaled ratio of the image to obtain the real size data of the glass to be inspected; The specific calculation formula for obtaining the real size data of the glass to be tested is: ; In the formula, The actual size data of the glass to be tested; The scaled size data of the glass to be inspected; is the image scaling ratio; In this embodiment, the actual size data of the glass to be inspected needs to be used as a reference to set the defect distance threshold and then make a quality judgment on the glass to be inspected. Therefore, the actual size data of the glass to be inspected is determined by calculating the scaling ratio of the image and the scaled size data of the glass to be inspected in the image.

[0023] Reference Figure 4As shown, based on the intelligent analysis terminal, feature analysis and processing are performed on the glass image to be detected, and the coordinates of the position of the defect of the glass to be detected are determined, which specifically includes the following steps: S301, analyzing and processing the image of the glass to be inspected based on the intelligent analysis terminal to determine the center position of the glass to be inspected; S302, based on the intelligent analysis terminal, the center position of the glass to be detected is set as the coordinate origin, the horizontal direction of the center position of the glass to be detected is set as the X-axis, and the vertical direction of the center position of the glass is set as the Y-axis, so as to construct a rectangular coordinate system; S303, based on the intelligent analysis terminal, feature analysis and processing are performed on the glass image to be detected to determine the position of the defect of the glass to be detected; S304: The intelligent analysis terminal places the position of the glass defect to be detected in a rectangular coordinate system to determine the coordinates of the position of the glass defect to be detected.

[0024] Reference Figure 5 As shown, based on the intelligent analysis terminal, the image of the glass to be detected is analyzed and processed to determine the center position of the glass to be detected, which specifically includes the following steps: S3011. Based on the intelligent analysis terminal, feature extraction and processing are performed on the glass image to be detected to determine the edge position of the glass to be detected; S3012, the intelligent analysis terminal draws a straight line from the edge position of the glass to be detected to the edge positions of other glasses to be detected, and obtains several groups of straight line intersection points; S3013, based on the intelligent analysis terminal, screening and processing a plurality of groups of straight line intersections to determine the center position of the glass to be tested; The specific calculation formula for determining the center position of the glass to be detected is:

[0025] In the formula, is the center position of the glass to be tested; is the number of lines that each straight line intersection passes through; n is the specific number of several groups of straight line intersections; max() is the maximum value function; In this embodiment, if the shape of the produced glass is irregular, it is more complicated to determine the center position of the glass, so a straight line is drawn from each edge feature of the irregularly shaped glass to other edge features, and the intersection points through which the straight line passes the most are the center positions of the glass to be detected.

[0026] Reference Figure 6 As shown, based on the intelligent analysis terminal, feature analysis and processing are performed on the glass image to be detected, and the location of the defect of the glass to be detected is determined, which specifically includes the following steps: S3031. Based on the intelligent analysis terminal, binarize the standard glass image and the glass image to be detected respectively to obtain the standard glass grayscale image and the glass grayscale image to be detected; S3032, based on the intelligent analysis terminal, respectively calculating the grayscale values ​​of the standard glass grayscale image and the glass grayscale image to be detected, and obtaining the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected; S3033. Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected are judged and processed to determine the position of the defect in the glass to be detected.

[0027] Reference Figure 7 As shown, based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected are judged and processed to determine the position of the defect of the glass to be detected, which specifically includes the following steps: S30331. Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be tested are judged; S30332. If the grayscale value of each pixel in the standard glass grayscale image is consistent with the grayscale value of each pixel in the grayscale image of the glass to be tested, the glass to be tested has no defects; S30333. If the grayscale value of each pixel in the standard glass grayscale image is inconsistent with the grayscale value of each pixel in the glass grayscale image to be detected, record the position where the grayscale value of the pixel is inconsistent, and set the position where the grayscale value of the pixel is inconsistent as the position of the glass defect to be detected; In this embodiment, when the firing process and raw materials of the glass are the same, and the images are obtained under the same shooting background and shooting parameters, if the glass does not have defects, then the grayscale values ​​of the pixels between the images are the same. Therefore, by comparing the grayscale value of each pixel in the standard glass image with the grayscale value of each pixel in the grayscale image of the glass to be tested, it is determined whether they are consistent. If they are inconsistent, it means that there are defects in the glass to be tested.

[0028] Reference Figure 8 As shown, based on the intelligent analysis terminal, the coordinates of the defect position of the glass to be detected are analyzed and processed to determine the quality of the glass to be detected, which specifically includes the following steps: S401, based on the intelligent analysis terminal, calculating and processing the position coordinates and coordinate origin of the glass defect to be detected, and determining the Euclidean distance of the glass defect to be detected; S402, extracting data from the database system based on the intelligent analysis terminal to obtain glass quality evaluation standards; S403, based on the intelligent analysis terminal, analyzing and processing the glass quality assessment standard and the real size data of the glass to be tested, and setting a defect distance threshold; S404, based on the intelligent analysis terminal, comparing and judging the defect distance threshold and the Euclidean distance of the glass defect to be detected, and determining the quality of the glass to be detected; The specific calculation formula of the Euclidean distance of the glass defect to be detected is:

[0029] In the formula, is the Euclidean distance of the glass defect to be detected; x is the X-axis coordinate of the position of the glass defect to be detected; y is the Y-axis coordinate of the position of the glass defect to be detected.

[0030] Reference Fig. 9 As shown, based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are compared and judged, and the quality of the glass to be detected is determined, which specifically includes the following steps: S4041. Based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are judged and processed; S4042, if the Euclidean distance of the glass defect to be detected is less than or equal to the first defect distance threshold, the quality of the glass to be detected is poor; S4043, if the Euclidean distance of the glass defect to be detected is greater than the first defect distance threshold, and the Euclidean distance of the glass defect to be detected is less than or equal to the second defect distance threshold, the quality of the glass to be detected is good; S4044, if the Euclidean distance of the defect of the glass to be detected is greater than the second threshold of the defect distance, the quality of the glass to be detected is excellent; In this embodiment, the defects in the glass may be far away from the center of the glass or near. If the defects are far away from the center of the glass, then a smaller portion of the defective glass needs to be cut to make the glass defect-free. The quality of the defective glass that has been cut is superior. Therefore, the quality of the defective glass to be inspected is determined by judging the distance between the defect in the glass to be inspected and the center of the glass to be inspected.

[0031] Reference Fig.10 As shown, a glass detection system based on artificial intelligence is used to implement the above-mentioned glass detection method based on artificial intelligence, including: An intelligent analysis terminal, which is used to control data transmission and information interaction between various modules; A database system, wherein the database system is used to store standard glass images and glass quality assessment standards; A parameter extraction module, wherein the parameter extraction module is used to perform parameter analysis on the standard glass image to determine the shooting parameters and shooting background of the standard glass image; A shooting device, wherein the shooting device is used to collect images of the glass to be inspected and obtain images of the glass to be inspected; A size calculation module, which performs calculations based on the scaled size data of the glass to be detected and the scaled ratio of the image to obtain the real size data of the glass to be detected; A coordinate system building module, wherein the coordinate system building module builds a rectangular coordinate system according to the center position of the glass to be detected; An image analysis module, which performs feature analysis on the image of the glass to be inspected to determine whether the glass to be inspected has defects; A data calculation module, which calculates and processes the position coordinates of the glass defect to be detected and the coordinate origin to determine the Euclidean distance of the glass defect to be detected; The quality judgment module compares and judges the defect distance threshold and the Euclidean distance of the defect of the glass to be detected to determine the quality of the glass to be detected.

[0032] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A glass detection method based on artificial intelligence, characterized in that: include: Based on the intelligent analysis terminal, control the shooting equipment to collect images of the glass and obtain the image of the glass to be tested; Based on the intelligent analysis terminal, the image of the glass to be inspected is analyzed and processed to obtain the real size data of the glass to be inspected; Based on the intelligent analysis terminal, the feature analysis and processing of the glass image to be inspected is performed to determine the position coordinates of the defects of the glass to be inspected; Based on the intelligent analysis terminal, the coordinates of the defect positions of the glass to be inspected are analyzed and processed to determine the quality of the glass to be inspected.

2. The glass detection method based on artificial intelligence according to claim 1, characterized in that: The method of controlling the shooting device to collect images of the glass based on the intelligent analysis terminal to obtain the image of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the database system is processed for information extraction to obtain standard glass images; Based on the intelligent analysis terminal, image parameter analysis and processing are performed on the standard glass image to obtain the shooting parameters and shooting background of the standard glass image; The intelligent analysis terminal adjusts the parameters of the shooting device according to the shooting parameters of the standard glass image to obtain the calibration shooting device; The glass to be inspected is placed in the shooting background, and based on the intelligent analysis terminal, the correction shooting equipment is controlled to collect images of the glass to be inspected to obtain the image of the glass to be inspected.

3. The glass detection method based on artificial intelligence according to claim 1, characterized in that: The method of analyzing and processing the image of the glass to be detected based on the intelligent analysis terminal to obtain the real size data of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the image of the glass to be inspected is measured and processed to obtain the scaled size data of the glass to be inspected; Based on the intelligent analysis terminal, the parameters of the calibration shooting equipment are extracted and processed to obtain the image scaling ratio; Based on the intelligent analysis terminal, the scaled size data of the glass to be inspected and the scaled ratio of the image are calculated and processed to obtain the real size data of the glass to be inspected; The specific calculation formula for obtaining the real size data of the glass to be tested is: ； In the formula, The actual size data of the glass to be tested; The scaled size data of the glass to be inspected; The image scaling factor.

4. The glass detection method based on artificial intelligence according to claim 3, characterized in that: The method of performing feature analysis on the glass image to be detected based on the intelligent analysis terminal to determine the position coordinates of the glass defect to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the image of the glass to be inspected is analyzed and processed to determine the center position of the glass to be inspected; Based on the intelligent analysis terminal, the center position of the glass to be detected is set as the coordinate origin, the horizontal direction of the center position of the glass to be detected is set as the X-axis, and the vertical direction of the center position of the glass is set as the Y-axis to construct a rectangular coordinate system; Based on the intelligent analysis terminal, the feature analysis and processing of the glass image to be inspected is performed to determine the location of the defect in the glass to be inspected; The intelligent analysis terminal places the position of the glass defect to be detected in a rectangular coordinate system to determine the coordinates of the position of the glass defect to be detected.

5. The glass detection method based on artificial intelligence according to claim 4, characterized in that: The method of analyzing and processing the image of the glass to be detected based on the intelligent analysis terminal to determine the center position of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, feature extraction and processing are performed on the glass image to be inspected to determine the edge position of the glass to be inspected; The intelligent analysis terminal draws a straight line from the edge position of the glass to be detected to the edge position of other glass to be detected, and obtains several groups of straight line intersection points; Based on the intelligent analysis terminal, several groups of straight line intersections are screened and processed to determine the center position of the glass to be tested; The specific calculation formula for determining the center position of the glass to be detected is: In the formula, is the center position of the glass to be tested; is the number of straight lines that each straight line intersection passes through; n is the specific number of several groups of straight line intersections; max() is the maximum value function.

6. The glass detection method based on artificial intelligence according to claim 4, characterized in that: The method of performing feature analysis on the glass image to be detected based on the intelligent analysis terminal to determine the position of the defect of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the standard glass image and the glass image to be detected are binarized to obtain the standard glass grayscale image and the glass grayscale image to be detected; Based on the intelligent analysis terminal, grayscale values ​​of the standard glass grayscale image and the glass grayscale image to be detected are calculated respectively to obtain the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected; Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected are judged and processed to determine the position of the defect in the glass to be detected.

7. The glass detection method based on artificial intelligence according to claim 6, characterized in that: The intelligent analysis terminal is used to judge and process the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be detected, and determine the position of the defect of the glass to be detected, which specifically includes the following steps: Based on the intelligent analysis terminal, the grayscale value of each pixel in the standard glass grayscale image and the grayscale value of each pixel in the glass grayscale image to be tested are judged; If the grayscale value of each pixel in the standard glass grayscale image is consistent with the grayscale value of each pixel in the grayscale image of the glass to be tested, the glass to be tested has no defects; If the grayscale value of each pixel in the standard glass grayscale image is inconsistent with the grayscale value of each pixel in the glass grayscale image to be detected, the position where the grayscale value of the pixel is inconsistent is recorded, and the position where the grayscale value of the pixel is inconsistent is set as the position of the glass defect to be detected.

8. The glass detection method based on artificial intelligence according to claim 1, characterized in that: The method of analyzing and processing the coordinates of the defect position of the glass to be detected based on the intelligent analysis terminal to determine the quality of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the coordinates of the position of the glass defect to be detected and the coordinate origin are calculated and processed to determine the Euclidean distance of the glass defect to be detected; Based on the intelligent analysis terminal, the database system is used to extract data and obtain glass quality evaluation standards; Based on the intelligent analysis terminal, the glass quality assessment standard and the actual size data of the glass to be tested are analyzed and processed to set the defect distance threshold; Based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are compared and judged to determine the quality of the glass to be detected; The specific calculation formula of the Euclidean distance of the glass defect to be detected is: In the formula, is the Euclidean distance of the glass defect to be detected; x is the X-axis coordinate of the position of the glass defect to be detected; y is the Y-axis coordinate of the position of the glass defect to be detected.

9. The glass detection method based on artificial intelligence according to claim 8, characterized in that: The method of comparing and judging the defect distance threshold and the Euclidean distance of the glass defect to be detected based on the intelligent analysis terminal to determine the quality of the glass to be detected specifically includes the following steps: Based on the intelligent analysis terminal, the defect distance threshold and the Euclidean distance of the glass defect to be detected are judged and processed; If the Euclidean distance of the glass defect to be detected is less than or equal to the first threshold of the defect distance, the quality of the glass to be detected is poor; If the Euclidean distance of the glass defect to be detected is greater than the first threshold of the defect distance, and the Euclidean distance of the glass defect to be detected is less than or equal to the second threshold of the defect distance, the quality of the glass to be detected is good; If the Euclidean distance of the defect in the glass to be detected is greater than the second threshold of the defect distance, the quality of the glass to be detected is excellent.

10. An artificial intelligence-based glass detection system, used to implement an artificial intelligence-based glass detection method as claimed in any one of claims 1 to 9, characterized in that: include: An intelligent analysis terminal, which is used to control data transmission and information interaction between various modules; A database system, wherein the database system is used to store standard glass images and glass quality assessment standards; A parameter extraction module, wherein the parameter extraction module is used to perform parameter analysis on the standard glass image to determine the shooting parameters and shooting background of the standard glass image; A shooting device, wherein the shooting device is used to collect images of the glass to be inspected and obtain images of the glass to be inspected; A size calculation module, which performs calculations based on the scaled size data of the glass to be detected and the scaled ratio of the image to obtain the real size data of the glass to be detected; A coordinate system building module, wherein the coordinate system building module builds a rectangular coordinate system according to the center position of the glass to be detected; An image analysis module, which performs feature analysis on the image of the glass to be inspected to determine whether the glass to be inspected has defects; A data calculation module, which calculates and processes the position coordinates of the glass defect to be detected and the coordinate origin to determine the Euclidean distance of the glass defect to be detected; The quality judgment module compares and judges the defect distance threshold and the Euclidean distance of the defect of the glass to be detected to determine the quality of the glass to be detected.

Images