Glass substrate defect detection device and detection method thereof

Abstract

The invention belongs to the technical field of defect detection, and particularly relates to defect detection of transparent materials, in particular to a glass substrate defect detection device and a detection method thereof.According to the glass substrate defect detection device, a containing groove is formed in a detection plate, liquid is added into the containing groove to form a water film, and a glass substrate is protected; the glass substrate is prevented from being scratched, meanwhile, the glass substrate cannot be lifted, interference to machine vision detection cannot be caused, meanwhile, two detection working conditions are set, accurate detection of scratches on the two faces of the to-be-detected glass substrate is completed through comparison of image information of the two detection working conditions, the scratches do not need to be avoided in the subsequent cutting process, and the detection efficiency is improved. And the surface where the scratches are located only needs to be thinned, so that the utilization rate of raw materials is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of defect detection, specifically relates to the defect detection of transparent materials, and particularly relates to a glass substrate defect detection device and a detection method thereof. Background Art

[0002] A glass substrate is an extremely flat thin glass sheet and is an important basic material in the field of semiconductor packaging. If there are scratches on the surface of the glass substrate, it will cause the insulation performance of the glass substrate to decline, leading to problems such as short circuits or leakage. Therefore, scratch detection of the glass substrate is a key step to ensure the reliability, yield, and performance of the product.

[0003] In the related art, a large glass substrate is cut, thinned, and polished to obtain a glass substrate of the required size. Before cutting, it is necessary to detect the surface scratches of the glass substrate. In order to avoid scratching the glass substrate on the detection table, many support blocks need to be set on the detection table to lift the glass substrate. However, during actual detection, due to the very strong light transmittance of the glass substrate (90%-95%), the lifted glass substrate and the detection table will cause interference to machine vision detection. When using machine vision detection, the scratches on the back of the glass substrate will be obtained by vision detection, and the surface where the scratches are located cannot be directly located. During subsequent cutting, only the method of avoiding the scratched area can be used for cutting, resulting in low raw material utilization rate.

[0004] Therefore, how to solve the interference caused by the support blocks to machine vision detection to improve the raw material utilization rate is a technical problem that needs to be solved urgently at present.

[0005] It should be noted that the above information disclosed in this background art part is only used to understand the background art of the concept of this application. Therefore, the above description is not considered as information of the prior art. Summary of the Invention

[0006] The embodiments of the present disclosure at least provide a glass substrate defect detection device and its working method.

[0007] In a first aspect, the embodiments of the present disclosure provide a glass substrate defect detection device, including: A detection platform; A detection board, which is arranged on the detection platform and is provided with a receiving groove for receiving the glass substrate to be detected; A driving component, which is arranged on the detection board and is used to drive the glass substrate to be detected into a first detection working condition and a second detection working condition; A liquid supply component, which is arranged on the detection platform and is used to inject liquid into the receiving groove to form a water film for protecting the glass substrate to be detected; A vision detection module, which is arranged above the detection plate and is used to obtain the image information of the glass substrate to be detected in the accommodation groove; A control module, which is configured to obtain the first image information of the glass substrate to be detected under the first detection condition and the second image information under the second detection condition through the vision detection module; And detect the scratches on both sides of the glass substrate to be detected according to the comparison result of the first image information and the second image information.

[0008] In a disclosed embodiment, the driving assembly includes: A pushing cylinder and two moving clamping members; The two moving clamping members are arranged oppositely and are arranged at both ends of the detection plate; The pushing cylinder is arranged at the bottom of the detection plate, and the piston rod of the pushing cylinder passes through the accommodation groove and is flush with the accommodation groove; Wherein, a liquid discharge hole is formed at the bottom of the accommodation groove, and a flange for blocking the liquid discharge hole extends radially outward from the top of the piston rod of the pushing cylinder; When the piston rod of the pushing cylinder extends out of the accommodation groove, the liquid discharge hole is opened to discharge the liquid.

[0009] In a disclosed embodiment, the control module controls the driving assembly to drive the glass substrate to be detected into the first detection condition, that is: Control the liquid supply assembly to inject liquid into the accommodation groove; Put the glass substrate into the accommodation groove through the suction cup of the external manipulator; Control the two moving clamping members to move towards each other to clamp the glass substrate to be detected, so that the glass substrate to be detected is in a horizontal state.

[0010] In a disclosed embodiment, the control module controls the driving assembly to drive the glass substrate to be detected into the second detection condition, that is: Control the two moving clamping members to move towards the pushing cylinder until one of the moving clamping members crosses the piston rod of the pushing cylinder, and then control the moving clamping member that crosses the piston rod of the pushing cylinder to continue moving a preset distance, and at the same time, control the other moving clamping member to stop moving; Control the pushing cylinder to lift one side of the glass substrate to be detected, so that the glass substrate to be detected is in an inclined state.

[0011] In a disclosed embodiment, when in the second detection condition, the preset inclination angle of the glass substrate to be detected is α; The range of α is between 20° and 50°; Among them, the extension distance L of the piston rod of the pushing cylinder is L = H * sinα; In the formula, L is the extension distance of the piston rod of the pushing cylinder, and H is the length of the glass substrate to be detected.

[0012] In an open embodiment, when in the first detection condition, the glass substrate to be detected is in a horizontal state; When in the second detection condition, the glass substrate to be detected is in an inclined state; The control module detects the scratches on both sides of the glass substrate to be detected according to the comparison result of the first image information and the second image information, that is: Preprocess the first image information and the second image information; Enlarge the second image information by K times along the length direction so that the edge position information of the first image information and the second image information is the same, where K is the magnification factor; Through the gray threshold segmentation method, obtain the first scratch pixel point coordinate set A1 of the first image information and the second scratch pixel point coordinate set A2 of the second image information according to the first gray threshold; Detect the scratches on both sides of the glass substrate to be detected according to the comparison result of the pixel point positions and the number of the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2.

[0013] In an open embodiment, among them, the calculation method of the magnification factor K is as follows: According to the edge detection algorithm, obtain the edge position information of the first image information and the second image information; Calculate the graphic length H1 of the second image information according to the edge position information; Calculate the ratio K = H / H1 of the length H of the glass substrate to be detected to the graphic length H1; Among them, the graphic length H1 refers to the length of the side corresponding to the long side of the glass substrate (600) in the second image information.

[0014] In an open embodiment, the control module detects the scratches on both sides of the glass substrate to be detected according to the comparison result of the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2, that is: First comparison result, if the first scratch pixel point coordinate set A1 is not an empty set, it means that there are scratches on the top surface of the glass substrate to be detected; Second comparison result, if the first scratch pixel point coordinate set A1 is not an empty set and A2 - A1 ≠ empty set, it means that there are scratches on the bottom surface of the glass substrate to be detected; Third comparison result, if A1 ∪ A2 = empty set, it means that there are no scratches on the glass substrate to be detected.

[0015] In an open embodiment, the control module is further configured to: When a scratch is detected on the glass substrate to be detected, control the liquid supply component to inject a staining solution into the accommodation groove; Then, after the scratched surface of the glass substrate to be detected faces the accommodation groove through an external vacuum suction cup manipulator, place it into the accommodation groove; Stain for a preset time; Obtain third image information of the glass substrate to be detected under the second detection condition through the vision detection module; Obtain the depth information of the scratch through the third image information for subsequent transfer and use.

[0016] In a second aspect, the embodiments of the present disclosure further provide a detection method applied to the glass substrate defect detection device as described above. The detection method includes: Inject a liquid into the accommodation groove through the liquid supply component to form a water film; Place the glass substrate to be detected into the accommodation groove; Control the driving component to drive the glass substrate to be detected into the first detection condition, and obtain first image information through the vision detection module; Control the driving component to drive the glass substrate to be detected into the second detection condition. At the same time, drain the liquid in the accommodation groove, and obtain second image information through the vision detection module after the liquid is drained; Detect scratches on both sides of the glass substrate to be detected based on the first image information and the second image information.

[0017] In an open embodiment, the detection method further includes: Screen the glass substrates with scratches; Control the liquid supply component to inject a staining solution into the accommodation groove; Then, after the scratched surface of the glass substrate to be detected faces the accommodation groove through an external vacuum suction cup manipulator, place it into the accommodation groove; Stain for a preset time; Control the driving component to drive the glass substrate to be detected into the second detection condition. At the same time, drain the liquid in the accommodation groove, and obtain third image information through the vision detection module after the liquid is drained; Obtain the depth information of the scratch through the third image information for subsequent transfer and use.

[0018] The beneficial effects of the present invention are as follows. The glass substrate defect detection device and its detection method of the present invention protect the glass substrate by providing a receiving groove on the detection plate and adding a liquid into the receiving groove to form a water film. While avoiding scratching the glass substrate, it will not lift the glass substrate, and thus will not interfere with machine vision detection. At the same time, two detection conditions are set, and through the comparison of the image information of the two detection conditions, the accurate detection of the scratches on both sides of the glass substrate to be detected is completed. During subsequent cutting, there is no need to avoid the scratches, and only the thinning of the surface where the scratches are located needs to be deepened, thereby improving the raw material utilization rate.

[0019] Other features and advantages of the present invention will be described in the subsequent description, and part of them will become obvious from the description, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the description and the drawings.

[0020] To make the above objectives, features, and advantages of the present invention more obvious and understandable, specific preferred embodiments are hereby given, and in conjunction with the accompanying drawings, the following detailed description is provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0022] Figure 1 Schematic structural diagram of the glass substrate defect detection device provided by an embodiment of the present disclosure; Figure 2 Schematic electrical control diagram of the glass substrate defect detection device provided by an embodiment of the present disclosure; Figure 3 Top view of the glass substrate defect detection device provided by an embodiment of the present disclosure; Figure 4 Front view of the glass substrate defect detection device provided by an embodiment of the present disclosure; Figure 5 Schematic diagram of the first image information provided by an embodiment of the present disclosure; Figure 6 Schematic diagram of the second image information and the enlarged second image information provided by an embodiment of the present disclosure; Figure 7 Comparison schematic diagram of the first image information and the second image information provided by an embodiment of the present disclosure; Figure 8 Flowchart of the detection method of the glass substrate defect detection device provided by an embodiment of the present disclosure.

[0023] In the figure: 100, detection platform; 200, detection board; 210, accommodation groove; 211, liquid discharge hole; 300, driving assembly; 310, pushing cylinder; 311, piston rod; 312, flanging; 320, moving clamping member; 321, moving plate; 322, driving cylinder; 400, liquid supply assembly; 500, vision detection module; 600, glass substrate. Detailed implementation manners

[0024] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] It has been found through research that when detecting a glass substrate in the related art, multiple support blocks need to be provided to lift the glass substrate to avoid scratching the glass substrate during detection. However, there is a large gap between the lifted glass substrate and the detection table. When using machine vision detection to detect the front surface of the glass substrate, scratches on the back surface will interfere with the detection, and the surface where the scratches are located cannot be accurately positioned. During subsequent cutting, only the method of avoiding the location where the scratches are located can be used for cutting, resulting in waste of raw materials.

[0026] Based on the above research, the embodiments of the present disclosure provide a glass substrate defect detection device and its detection method. By filling the space between the glass substrate and the detection board with a water film, the glass substrate can be prevented from being scratched during detection, and at the same time, it will not interfere with machine vision detection. The surface where the scratches are located can be accurately detected, which is convenient for removing the scratches during subsequent polishing, thereby improving the utilization rate of raw materials.

[0027] Regarding the defects existing in the above solutions, they are all the results obtained by the inventor through practice and careful research. Therefore, the process of discovering the above problems and the solutions proposed by the present disclosure in this article for the above problems should be the contributions made by the inventor to the present disclosure during the process of the present disclosure.

[0028] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0029] The following will describe in detail some embodiments of the present invention with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0030] Please refer toFigure 1 and Figure 2 At least one embodiment provides a glass substrate defect detection device, including the following structures: A detection platform 100.

[0031] A detection plate 200, which is arranged on the detection platform 100 and is provided with a receiving groove 210 for receiving a glass substrate 600 to be detected. Among them, the bottom of the receiving groove 210 is a black matte ceramic plate to reduce the reflection of light by the detection plate 200, and its surface roughness is less than 0.3μm, so as to avoid the roughness of the detection plate 200 interfering with the scratch detection.

[0032] A driving component 300, which is arranged on the detection plate 200 and is used to drive the glass substrate 600 to be detected into the first detection working condition and the second detection working condition.

[0033] A liquid supply component 400, which is arranged on the detection platform 100 and is used to inject liquid into the receiving groove 210 to form a water film for protecting the glass substrate 600 to be detected.

[0034] Specifically, the liquid supply component 400 includes at least one liquid spraying head, and the liquid sprayed by the liquid spraying head is pure water.

[0035] By arranging the receiving groove 210 on the detection plate 200 and adding liquid to the receiving groove 210 to form a water film to protect the glass substrate 600, while avoiding scratching the glass substrate 600, it will not lift the glass substrate 600, and thus will not interfere with the machine vision detection.

[0036] A vision detection module 500, which is arranged above the detection plate 200 and is used to obtain image information of the glass substrate 600 to be detected in the receiving groove 210.

[0037] Among them, the vision detection module 500 is a 5-million-pixel industrial camera, with a pixel size of 2μm and a frame rate of 240fps.

[0038] A control module, which is configured to obtain first image information of the glass substrate 600 to be detected in the first detection working condition and second image information in the second detection working condition through the vision detection module 500; and detect scratches on both sides of the glass substrate 600 to be detected according to the comparison result of the first image information and the second image information. By setting two detection working conditions and comparing the image information of the two detection working conditions, accurate detection of scratches on both sides of the glass substrate 600 to be detected is completed. During subsequent cutting, there is no need to avoid scratches, and only the thinning of the surface where the scratches are located needs to be deepened, thereby improving the raw material utilization rate.

[0039] Please refer toFigure 3 and Figure 4 The driving assembly 300 includes a pushing cylinder 310 and two moving clamping members 320; the two moving clamping members 320 are arranged oppositely and are arranged at both ends of the detection plate 200; the pushing cylinder 310 is arranged at the bottom of the detection plate 200, and the piston rod 311 of the pushing cylinder 310 passes through the receiving groove 210 and is flush with the receiving groove 210.

[0040] The position of the glass substrate 600 is adjusted by driving the moving clamping assembly through the control module, and at the same time, the pushing of the pushing cylinder 310 is supplemented to enable the glass substrate 600 to enter the first detection working condition and the second detection working condition.

[0041] Wherein, a liquid discharge hole 211 is formed at the bottom of the receiving groove 210, and a flange 312 for blocking the liquid discharge hole 211 extends radially outward from the top of the piston rod 311 of the pushing cylinder 310; when the piston rod 311 of the pushing cylinder 310 extends out of the receiving groove 210, the liquid discharge hole 211 is opened to discharge the liquid.

[0042] In the second detection working condition, the liquid discharge hole 211 is opened, so as to facilitate the discharge of the liquid in the receiving groove 210.

[0043] The following is an explanation of the first working condition and the second working condition: The control module controls the driving assembly 300 to drive the glass substrate 600 to be detected into the first detection working condition, that is: Controlling the liquid supply assembly 400 to inject liquid into the receiving groove 210; placing the glass substrate 600 into the receiving groove 210 through the suction cup of the external manipulator; controlling the two moving clamping members 320 to move towards each other to clamp the glass substrate 600 to be detected, so that the glass substrate 600 to be detected is in a horizontal state.

[0044] A schematic diagram of the glass substrate 600 under the first detection working condition is as Figure 3 shown. At this time, the two moving clamping members 320 clamp the glass substrate 600.

[0045] It should be noted that the thickness of the glass substrate 600 is greater than the depth of the receiving groove 210, so as to facilitate the clamping of the glass substrate 600 by the moving clamping members 320.

[0046] The control module controls the driving assembly 300 to drive the glass substrate 600 to be detected into the second detection working condition, that is: Controlling the two moving clamping members 320 to move towards the pushing cylinder 310 (the moving direction is as Figure 3As shown in F in the figure, until one of the moving clamping members 320 passes over the piston rod 311 of the pushing cylinder 310, control the moving clamping member 320 that has passed over the piston rod 311 of the pushing cylinder 310 to continue moving a preset distance. At the same time, control the other moving clamping member 320 to stop moving; control the pushing cylinder 310 to lift one side of the glass substrate 600 to be detected, so that the glass substrate 600 to be detected is in an inclined state.

[0047] A schematic diagram of the glass substrate 600 under the second detection condition is as Figure 4 shown. At this time, the pushing cylinder 310 pushes upward along the Figure 4 direction shown by F in the figure, so that the glass substrate 600 to be detected is inclined.

[0048] Among them, the moving clamping member 320 includes a moving plate 321 and a driving cylinder 322. The driving cylinder 322 drives the moving plate 321 to slide on the detection plate 200 to complete the adjustment of the position of the glass substrate 600.

[0049] Please continue to refer to Figure 4 , when in the second detection condition, the preset inclination angle of the glass substrate 600 to be detected is α; the range of α is between 20° - 50°; among them, the extension distance L of the piston rod 311 of the pushing cylinder 310 = H * sinα; in the formula, L is the extension distance of the piston rod 311 of the pushing cylinder 310, and H is the length of the glass substrate 600 to be detected. By inclining the glass substrate 600, subsequent detection of scratches on the back surface of the glass substrate 600 is carried out.

[0050] In a preferred embodiment, α is 45°, so that the scratches on the back surface and the front surface are fully misaligned, facilitating subsequent scratch detection.

[0051] It should be noted that the preset inclination angle is the set angle, not the actual inclination angle of the glass substrate 600. The function of this angle is to confirm the extension distance of the piston rod 311 of the pushing cylinder 310.

[0052] Please continue to refer to Figure 3 and Figure 4 , when in the first detection condition, the glass substrate 600 to be detected is in a horizontal state; when in the second detection condition, the glass substrate 600 to be detected is in an inclined state.

[0053] The control module detects the scratches on both sides of the glass substrate 600 to be detected according to the comparison result of the first image information and the second image information, that is: Step S1, preprocess the first image information and the second image information.

[0054] Specifically, the collected color image is converted into a grayscale image, and then the grayscale image is denoised by Gaussian filtering. During this process, the interference caused by the water film to the image is eliminated.

[0055] Step S2: Magnify the second image information by K times along the length direction so that the edge position information of the first image information and the second image information is the same, where K is the magnification factor.

[0056] Specifically, since the second image information is taken after the glass substrate 600 is tilted, at this time, the contour of the glass substrate 600 in the obtained second image information is shorter than the contour of the glass substrate 600 in the first image information. It is necessary to magnify the second image information along the length direction according to a certain ratio before the contours of the glass substrate 600 corresponding to the first image information and the second image information are of the same size, so as to facilitate subsequent comparison.

[0057] Among them, the schematic diagram of the first image information is as Figure 5 shown, Figure 5 where a and b in Figure 6 are two scratches existing on the top surface of the glass substrate 600; the schematic diagrams of the second image information and the magnified second image information are as Figure 6 shown,

[0058] where a and b in

[0059] are two scratches existing on the top surface of the glass substrate 600, c and d are two scratches existing on the bottom surface of the glass substrate 600, a' and b' are two scratches existing on the top surface of the glass substrate 600 in the magnified second image information, and c' and d' are two scratches existing on the bottom surface of the glass substrate 600 in the magnified second image information.

[0060] Among them, A is the average value of the first image information and the second image information obtained by the scratch-free glass substrate 600 under the first detection condition and the second detection condition, and the first gray threshold T = β * A.

[0061] Among them, T is the first gray threshold; β is a compensation coefficient, and β < 0.9.

[0062] For example, in a certain example, A1 = {(100, 99), (101, 100), (101, 101), (102, 101), (102, 102), (102, 103), (102, 104)}, A2 = {(100, 99), (101, 100), (101, 101), (102, 101), (102, 102), (102, 103), (102, 104), (200, 99), (201, 99), (202, 99)}, where (100, 99) indicates that the gray value of the pixel at the 100th row and 99th column is less than the first gray threshold, and so on. (101, 100) indicates that the gray value of the pixel at the 101st row and 100th column is less than the first gray threshold. Similarly, (101, 101)…

[0063] It should be noted that here is just a simple explanation of the data type structure of the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2, and does not represent the number of pixel points in the actual detection process. Step S4: Detect the scratches on both sides of the glass substrate 600 to be detected according to the comparison result between the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2.

[0064] Specifically, for the first comparison result, if the first scratch pixel point coordinate set A1 is not an empty set, it means that there is a scratch on the top surface of the glass substrate 600 to be detected. As Figure 5 shown.

[0065] For the second comparison result, if the first scratch pixel point coordinate set A1 is not an empty set and A2 - A1 ≠ empty set, it means that there is a scratch on the bottom surface of the glass substrate 600 to be detected. As Figure 7 shown. In the above example, A2 - A1 = {(200, 99), (201, 99), (202, 99)}, and (200, 99), (201, 99), (202, 99) are the position coordinates of the scratches on the bottom surface.

[0066] For the third comparison result, if A1 ∪ A2 = empty set, it means that there is no scratch on the glass substrate 600 to be detected. That is, if there are no pixel points less than the first gray threshold in both A1 and A2, it indicates that there is no scratch on the glass substrate 600 to be detected.

[0067] It should be noted that the calculation method of the magnification factor K is as follows: According to the edge detection algorithm, obtain the edge position information of the first image information and the second image information.

[0068] Calculate the graphic length H1 of the second image information according to the edge position information.

[0069] Calculate the ratio K = H / H1 of the length H of the glass substrate 600 to be detected to the pattern length H1, where the pattern length H1 refers to the length of the side corresponding to the long side of the glass substrate (600) in the second image information.

[0070] The control module is further configured to: when a scratch is detected on the glass substrate 600 to be detected, control the liquid supply assembly 400 to inject a staining solution into the receiving groove 210; then, after the scratched surface of the glass substrate 600 to be detected faces the receiving groove 210, place it into the receiving groove 210; stain for a preset time (20 minutes); obtain the third image information of the glass substrate 600 to be detected under the second detection condition through the vision detection module 500; obtain the depth information of the scratch through the third image information for subsequent transfer and use.

[0071] Specifically, when detecting the depth of the scratch on the scratched surface, the staining method is used for detection. According to experimental data, when the scratch is between 5 - 50 μm, the color depth of the scratch is linearly correlated with the 0.8th power of the scratch volume.

[0072] The depth is calculated by the following formula: RGBavg = (R + G + B) / 3 = mV 0.8 +b; In the formula, RGBavg represents the average color value of the scratch; R represents the color value of the red channel; G represents the color value of the green channel; B represents the color value of the blue channel; m represents the exponential fitting coefficient, ranging from 0.5 - 2; b is the background color value of the detection plate 200, ranging from 50 - 150; V represents the scratch volume (unit: μm 3 )

[0073] The length (unit: μm) l and width w (unit: μm) of the scratch are obtained through image analysis, and then the scratch depth h (unit: μm) is further calculated. The formula is as follows: h = V / (wl).

[0074] Assume that the thickness of the glass substrate 600 is W 总 (unit: μm), and the thickness of the thinned glass substrate 600 is W 实 (unit: μm), and the polishing wear thickness is λ (unit: μm); h < W 总 -W 实 -2λ indicates that the scratch will not affect the normal production of the glass substrate 600 and can meet the subsequent thinning and polishing requirements.

[0075] It should be noted that the scratch detection is divided into two stages. In the first stage, pure water is used to form a water film for the initial screening of the glass substrate 600. After the glass substrate 600 with scratches is screened out, it is centrally subjected to a secondary detection to detect the scratch depth. During the secondary detection, a staining agent is sprayed through another spray head of the liquid supply assembly 400.

[0076] Please refer to Figure 8 , the present disclosure embodiment also provides a detection method applied to the glass substrate defect detection device as described above. The detection method includes: S110: Inject a liquid into the receiving groove 210 through the liquid supply assembly 400 to form a water film.

[0077] S120: Place the glass substrate 600 to be detected into the receiving groove 210.

[0078] S130: Control the driving assembly 300 to drive the glass substrate 600 to be detected into the first detection working condition, and obtain the first image information through the vision detection module 500.

[0079] S140: Control the driving assembly 300 to drive the glass substrate 600 to be detected into the second detection working condition. At the same time, drain the liquid in the receiving groove 210, and obtain the second image information through the vision detection module 500 after the liquid is drained.

[0080] S150: Detect the scratches on both sides of the glass substrate 600 to be detected based on the first image information and the second image information.

[0081] Steps S110 - S150 are preliminary detections to screen out the glass substrates 600 with scratches. By setting the receiving groove 210 on the detection plate 200 and adding a liquid into the receiving groove 210 to form a water film, the glass substrate 600 is protected. While avoiding scratching the glass substrate 600, it will not lift the glass substrate 600, and thus will not interfere with the machine vision detection. At the same time, two detection working conditions are set, and through the comparison of the image information of the two detection working conditions, the accurate detection of the scratches on both sides of the glass substrate 600 to be detected is completed. During subsequent cutting, there is no need to avoid the scratches, and only the thinning of the surface where the scratches are located needs to be deepened, thereby improving the raw material utilization rate.

[0082] After screening out the glass substrates 600 with scratches, the detection method further includes: S210: Screen the glass substrates 600 with scratches.

[0083] S220: Control the liquid supply assembly 400 to inject a staining liquid into the receiving groove 210.

[0084] S230: Then, after the scratched surface of the glass substrate 600 to be detected faces the receiving groove 210, it is placed into the receiving groove 210 by an external vacuum suction cup manipulator.

[0085] S240: Dye for a preset time.

[0086] S250: Control the driving component 300 to drive the glass substrate 600 to be detected into the second detection condition. At the same time, drain the liquid in the receiving groove 210, and after the liquid is drained, obtain the third image information through the vision detection module 500. S260: Obtain the depth information of the scratch through the third image information for subsequent transfer and use.

[0087] By detecting the depth of the scratch, it is judged whether the scratch depth h meets the usage requirements of subsequent thinning and polishing. When the requirements of thinning and polishing are met, only the thinning of the surface where the scratch is located needs to be deepened, thereby improving the raw material utilization rate.

[0088] In summary, the present invention provides a glass substrate defect detection device and its detection method. By setting a receiving groove 210 on the detection plate 200 and adding a liquid to form a water film in the receiving groove 210, the glass substrate 600 is protected. While avoiding scratching the glass substrate 600, it will not lift the glass substrate 600, and thus will not interfere with machine vision detection. At the same time, two detection conditions are set, and through the comparison of the image information of the two detection conditions, the accurate detection of the scratches on both sides of the glass substrate 600 to be detected is completed. During subsequent cutting, there is no need to avoid the scratches, and only the thinning of the surface where the scratches are located needs to be deepened, thereby improving the raw material utilization rate.

[0089] Enlightened by the above ideal embodiments of the present invention, through the above description, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A glass substrate defect detection device, characterized in that: include: Detection platform (100); A detection plate (200), which is arranged on the detection platform (100) and is provided with a receiving groove (210) for receiving a glass substrate (600) to be detected; A driving component (300) disposed on the detection plate (200) and used to drive the glass substrate (600) to be detected to enter a first detection working state and a second detection working state; a liquid supply component (400), which is arranged on the detection platform (100) and is used to inject liquid into the containing groove (210) to form a water film for protecting the glass substrate (600) to be detected; a visual inspection module (500), which is arranged above the inspection plate (200) and is used to obtain image information of the glass substrate (600) to be inspected in the receiving groove (210); A control module configured to obtain, through the visual inspection module (500), first image information of a glass substrate (600) to be inspected under a first inspection condition, and second image information under a second inspection condition; The scratches on both sides of the glass substrate (600) to be inspected are inspected based on the comparison results of the first image information and the second image information.

2. The glass substrate defect detection device according to claim 1, characterized in that: The driving assembly (300) comprises: A push cylinder (310) and two movable clamping members (320); The two movable clamping members (320) are arranged opposite to each other and are arranged at two ends of the detection plate (200); The push cylinder (310) is arranged at the bottom of the detection plate (200), and the piston rod (311) of the push cylinder (310) passes through the receiving groove (210) and is flush with the receiving groove (210); A drainage hole (211) is provided at the bottom of the containing groove (210), and a flange (312) for sealing the drainage hole (211) is radially extended outward from the top of the piston rod (311) of the push cylinder (310); When the piston rod (311) of the push cylinder (310) extends out of the containing groove (210), the liquid discharge hole (211) is opened to discharge the liquid.

3. The glass substrate defect detection device according to claim 2, characterized in that: The control module controls the driving component (300) to drive the glass substrate (600) to be inspected to enter a first inspection state, namely: Controlling the liquid supply component (400) to inject liquid into the containing tank (210); Placing the glass substrate (600) into the containing groove (210) by means of a suction cup of an external robot; The two movable clamping members (320) are controlled to move towards each other to clamp the glass substrate (600) to be inspected, so that the glass substrate (600) to be inspected is in a horizontal state.

4. The glass substrate defect detection device according to claim 2, wherein: The control module controls the driving component (300) to drive the glass substrate (600) to be inspected to enter a second inspection state, namely: Controlling the two movable clamping members (320) to move toward the push cylinder (310) until one of the movable clamping members (320) passes over the piston rod (311) of the push cylinder (310), and then controlling the movable clamping member (320) that passes over the piston rod (311) of the push cylinder (310) to continue to move a preset distance, while controlling the other movable clamping member (320) to stop moving; The push cylinder (310) is controlled to push up one side of the glass substrate (600) to be inspected, so that the glass substrate (600) to be inspected is in a tilted state.

5. The glass substrate defect detection device according to claim 4, characterized in that: In the second detection working condition, the preset tilt angle of the glass substrate (600) to be detected is α; The range of α is between 20°-50°; Wherein, the extension distance of the piston rod (311) of the push cylinder (310) is L=H*sinα; In the formula, L is the extension distance of the piston rod (311) of the push cylinder (310), and H is the length of the glass substrate (600) to be inspected.

6. The glass substrate defect detection device according to claim 1, wherein: In the first detection working condition, the glass substrate (600) to be detected is in a horizontal state; In the second detection working condition, the glass substrate (600) to be detected is in a tilted state; The control module detects scratches on both sides of the glass substrate (600) to be detected based on the comparison results of the first image information and the second image information, namely: Preprocessing the first image information and the second image information; Enlarging the second image information by K times along the length direction so that the edge position information of the first image information and the second image information are the same, wherein K is the enlargement factor; By using a grayscale threshold segmentation method, a first scratch pixel point coordinate set A1 of the first image information and a second scratch pixel point coordinate set A2 of the second image information are obtained according to a first grayscale threshold; According to the comparison results of the positions and numbers of the pixel points of the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2, the scratches on both sides of the glass substrate (600) to be inspected are inspected.

7. The glass substrate defect detection device according to claim 6, characterized in that: in, The magnification K is calculated as follows: Acquire edge position information of the first image information and the second image information according to an edge detection algorithm; Calculating a graphic length H1 of the second image information according to the edge position information; Calculating a ratio K=H / H1 of a length H of the glass substrate (600) to be inspected and a length H1 of the pattern; The graphic length H1 refers to the length of the side corresponding to the long side of the glass substrate (600) in the second image information.

8. The glass substrate defect detection device according to claim 6, wherein: The control module detects scratches on both sides of the glass substrate (600) to be detected based on the comparison result of the first scratch pixel point coordinate set A1 and the second scratch pixel point coordinate set A2, that is: The first comparison result indicates that the first scratch pixel point coordinate set A1 is not an empty set, which indicates that there is a scratch on the top surface of the glass substrate (600) to be inspected; The second comparison result shows that the first scratch pixel point coordinate set A1 is not an empty set, and A2-A1≠empty set, indicating that there is a scratch on the bottom surface of the glass substrate (600) to be inspected; The third comparison result, A1∪A2=empty set, indicates that the glass substrate (600) to be inspected does not have any scratches.

9. The glass substrate defect detection device according to claim 1, wherein: The control module is further configured to: When it is detected that the glass substrate (600) to be inspected has scratches, controlling the liquid supply component (400) to inject dyeing liquid into the containing tank (210); Then, the glass substrate (600) to be inspected is placed in the receiving groove (210) with the scratched surface facing the receiving groove (210) by an external vacuum suction cup robot; Dyeing preset time; Acquiring third image information of a glass substrate (600) to be inspected under a second inspection condition through a visual inspection module (500); The depth information of the scratch is obtained through the third image information for subsequent circulation.

10. A detection method applied to the glass substrate defect detection device according to any one of claims 1 to 9, characterized in that: The detection method comprises: Injecting liquid into the containing tank (210) through the liquid supply component (400) to form a water film; Placing a glass substrate (600) to be inspected into the receiving tank (210); Controlling the driving component (300) to drive the glass substrate (600) to be inspected to enter a first inspection working state, and obtaining first image information through the visual inspection module (500); Controlling the driving component (300) to drive the glass substrate (600) to be inspected to enter a second inspection state, and at the same time, draining the liquid in the containing tank (210), and obtaining second image information through the visual inspection module (500) after the liquid is drained; Scratches on both sides of a glass substrate (600) to be inspected are inspected based on the first image information and the second image information.

11. The detection method of the glass substrate defect detection device according to claim 10, characterized in that: The detection method further comprises: Screening the glass substrate (600) with scratches; Controlling the liquid supply component (400) to inject dyeing liquid into the containing tank (210); Then, the glass substrate (600) to be inspected is placed in the receiving groove (210) with the scratched surface facing the receiving groove (210) by an external vacuum suction cup robot; Dyeing preset time; Controlling the driving component (300) to drive the glass substrate (600) to be inspected to enter a second inspection state, and at the same time, draining the liquid in the containing tank (210), and obtaining third image information through the visual inspection module (500) after the liquid is drained; The depth information of the scratch is obtained through the third image information for subsequent circulation.

Images