A screen inspection management system for LCD production line based on machine vision

Abstract

The present invention belongs to the field of screen detection and relates to data analysis technology, and is used to solve the problems of complicated detection procedures and low efficiency of screen detection management systems for liquid crystal display production lines in the prior art. Specifically, the present invention is a screen detection management system for liquid crystal display production lines based on machine vision, and a gradient test unit is used to perform gradient display test analysis on the screen of the liquid crystal display: the screen of the liquid crystal display is marked as a detection object, the detection object is divided into a plurality of detection areas, a horizontal data group and a vertical data group of the detection area are obtained, and the horizontal data group and the vertical data group are sent to the gradient analysis unit; the present invention can perform gradient display test analysis on the screen of the liquid crystal display, limit the grayscale distribution characteristics of the pixel grid of the detection area through the equipped test image, and then evaluate multiple performances of the detection area simultaneously according to the detection data of the horizontal data group, the vertical data group and the machine vision group, so as to improve the detection efficiency.

Description

Technical Field

[0001] The invention belongs to the field of screen detection and relates to data analysis technology, and specifically is a screen detection management system for a liquid crystal display production line based on machine vision. Background Art

[0002] Liquid crystal display is a very important component in electronic products. It is widely used in various devices such as mobile phones, televisions, computers, etc. To ensure the quality and normal operation of liquid crystal display, strict testing is essential.

[0003] The screen inspection and management system of the LCD display production line in the prior art cannot simultaneously inspect the gradient display performance, pure color display performance and angle display consistency of the screen, resulting in cumbersome inspection procedures and low efficiency.

[0004] In view of the above technical problems, this application proposes a solution. Summary of the invention

[0005] The purpose of the present invention is to provide a liquid crystal display production line screen detection management system based on machine vision, which is used to solve the problem of complicated detection process and low efficiency of the liquid crystal display production line screen detection management system in the prior art;

[0006] The technical problem to be solved by the present invention is: how to provide a screen detection management system for a liquid crystal display production line based on machine vision with a concise detection process.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A screen detection management system for a liquid crystal display production line based on machine vision, comprising a gradient test unit, a gradient analysis unit and a data processing unit;

[0009] The gradient test unit is used to perform a gradient display test analysis on the screen of the liquid crystal display screen: mark the screen of the liquid crystal display screen as a detection object, divide the detection object into a plurality of detection areas, mark the horizontal pixel grids of the detection area as horizontal grids i, i=1, 2, ..., n, n is a positive integer, mark the vertical pixel grids of the detection area as vertical grids e, e=1, 2, ..., m, m is a positive integer, and generate a test image through software; display the test image in the detection area of ​​the detection object, and then randomly select a camera from the machine vision group to capture the image of the detection area, the machine vision group includes cameras set at different angles, and the captured images are processed to obtain m horizontal data groups and n vertical data groups, and the horizontal data groups and the vertical data groups are sent to the gradient analysis unit;

[0010] The gradient analysis unit is used to analyze the gradient performance of the liquid crystal display screen and obtain a variable coefficient BY of the detection area, and mark the detection area as a normal area or an abnormal area through the variable coefficient BY;

[0011] The data processing unit is used to process and analyze the detection data of the liquid crystal display screen.

[0012] Furthermore, the pixel grids of the test image meet the following requirements: the pixel grids of the test image are distributed in m rows and n columns, the grayscale value of the pixel grid in the first column of the test image is zero, the grayscale value of the pixel grid in the second column is 0+255 / (n-1), the grayscale value of the pixel grid in the third column is 0+255 / (n-1)×2, and so on, and the grayscale value of the pixel grid in the last column is 255.

[0013] Furthermore, the specific process of processing the captured image includes: enlarging the captured image into a pixel grid image and performing grayscale transformation, and then obtaining m horizontal data groups and n vertical data groups, the process of obtaining the horizontal data group includes: selecting a row of pixel grids as processing grids, marking the absolute value of the difference between the second pixel grid grayscale value and the first pixel grid grayscale value as the first processing value, marking the absolute value of the difference between the third pixel grid grayscale value and the second pixel grid grayscale value as the second processing value, and so on, marking the absolute value of the difference between the nth pixel grid and the n-1th pixel grid grayscale value as the n-1th processing value, and the n-1 processing values ​​constitute a horizontal data group; the process of obtaining the vertical data group includes: selecting a column of pixel grids as marking grids, and the m marking grids constitute a vertical data group.

[0014] Furthermore, the process of obtaining the variable coefficient BY of the detection area includes: performing variance calculation on all elements in the horizontal data group to obtain the horizontal coefficient of the horizontal data group, summing and averaging the horizontal coefficients of m horizontal data groups to obtain the horizontal performance value HB; performing variance calculation on all elements in the vertical data group to obtain the vertical coefficient of the vertical data group, summing and averaging the vertical coefficients of n vertical data groups to obtain the vertical performance value ZB; and obtaining the variable coefficient BY of the detection area by performing numerical calculation on the horizontal performance value HB and the vertical performance value ZB.

[0015] Furthermore, the specific process of marking the detection area as a normal area or an abnormal area includes: comparing the variable coefficient BY of the detection area with the preset variable threshold BYmax: if the variable coefficient BY is less than the variable threshold BYmax, the corresponding detection area is marked as a normal area; if the variable coefficient BY is greater than or equal to the variable threshold BYmax, the corresponding detection area is marked as an abnormal area.

[0016] Furthermore, the specific process of the data processing unit processing and analyzing the detection data of the liquid crystal display screen includes: marking the ratio of the number of abnormal areas in the detection object to the number of detection areas as the processing coefficient CL of the detection object, and comparing the processing coefficient CL of the detection object with the preset processing thresholds CLmin and CLmax:

[0017] If CL≤CLmin, it is determined that the gradient consistency of the processed object meets the requirements, and the corresponding processed object is marked as a qualified object;

[0018] If CLmin<CL<CLmax, it is determined that the consistency of the processing object does not meet the requirements, the corresponding processing object is marked as a consistent optimization object, a consistent abnormal signal is generated and sent to the mobile terminal of the manager;

[0019] If CL≥CLmax, it is determined that the gradual change of the processing object does not meet the requirements, the corresponding processing object is marked as a gradual change optimization object, a gradual change optimization signal is generated and sent to the mobile phone terminal of the manager.

[0020] The present invention has the following beneficial effects:

[0021] 1. The gradient test unit can be used to perform gradient display test analysis on the LCD screen, monitor the image display status of each detection area in a regional analysis manner, define the pixel grid grayscale distribution characteristics of the detection area through the equipped test image, and then evaluate multiple performances of the detection area simultaneously based on the detection data of the horizontal data group, vertical data group and machine vision group, thereby improving the detection efficiency;

[0022] 2. The gradient analysis unit can be used to analyze the gradient performance of the LCD screen. After processing the horizontal data group and the vertical data group, the variable coefficient is obtained. According to the variable coefficient, the gradient display performance and the pure color display performance of the LCD screen are fed back. Thus, the detection area is differentially marked by the variable coefficient, providing data support for the overall screen performance analysis;

[0023] 3. The data processing unit can process and analyze the test data of the LCD screen, and provide feedback on the overall performance of the LCD screen based on the number and proportion of abnormal areas in the test object, so as to sort and rework the test objects with unqualified consistency and unqualified gradient, thereby improving the quality of the LCD screen;

[0024] 4. The calibration and analysis module can be used to perform calibration and analysis on the machine vision group. The calibration data can be cleaned and processed regularly to obtain the abnormal calibration value of each camera. Feedback on the necessity of camera calibration can be provided based on the abnormal calibration value to avoid abnormal camera operation that may lead to a decrease in the accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 A system block diagram of the present invention as a whole;

[0027] Figure 2 is a system block diagram of Embodiment 1 of the present invention;

[0028] Figure 3 is a system block diagram of Embodiment 2 of the present invention;

[0029] Figure 4 This is a method flow chart of embodiment 3 of the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] like Figure 1 As shown, a screen detection and management system for a liquid crystal display production line based on machine vision includes a detection and analysis module and a verification and analysis module. The detection and analysis module includes a gradient test unit, a gradient analysis unit and a data processing unit; the verification and analysis module includes a data cleaning unit and a verification processing unit.

[0032] Embodiment 1

[0033] like Figure 2 As shown, the gradual change test unit, the gradual change analysis unit and the data processing unit are communicatively connected in sequence.

[0034] The gradient test unit is used to perform gradient display test analysis on the screen of the liquid crystal display: the screen of the liquid crystal display is marked as a detection object, the detection object is divided into several detection areas, the horizontal pixel grids of the detection area are marked as horizontal grids i, i = 1, 2, ..., n, n is a positive integer, and the vertical pixel grids of the detection area are marked as vertical grids e, e = 1, 2, ..., m, m is a positive integer, and a test image is generated by software. The pixel grids of the test image meet the following requirements: the pixel grids of the test image are distributed according to m rows and n columns, and the first pixel grid of the test image is The grayscale value of the first pixel grid is zero, the grayscale value of the second pixel grid is 0+255 / (n-1), the grayscale value of the second pixel grid is 0+255 / (n-1)×2, and so on. The grayscale value of the last pixel grid is 255. The test image is displayed in the detection area of ​​the detection object, and then a camera is randomly selected from the machine vision group to capture the image of the detection area. The machine vision group includes cameras set at different angles. The captured image is enlarged into a pixel grid image and grayscale transformed, and then m horizontal data are obtained. The method comprises the following steps: selecting a row of pixel grids as a processing grid, marking the absolute value of the difference between the grayscale value of the second pixel grid and the grayscale value of the first pixel grid as the first processing value, marking the absolute value of the difference between the grayscale value of the third pixel grid and the grayscale value of the second pixel grid as the second processing value, and so on, marking the absolute value of the difference between the grayscale value of the nth pixel grid and the n-1th pixel grid as the n-1th processing value, and forming a horizontal data group with n-1 processing values; the method comprises the following steps: selecting a column of pixel grids as a marking grid, and forming a vertical data group with m marking grids; sending the horizontal data group and the vertical data group to a gradient analysis unit; performing a gradient display test analysis on the screen of the liquid crystal display screen, monitoring the image display state of each detection area in a regional analysis manner, limiting the grayscale distribution characteristics of the pixel grids in the detection area through the provided test image, and then simultaneously evaluating multiple performances of the detection area according to the detection data of the horizontal data group, the vertical data group and the machine vision group, so as to improve the detection efficiency.

[0035] The gradient analysis unit is used to analyze the gradient performance of the liquid crystal display screen: the variance of all elements in the horizontal data group is calculated to obtain the horizontal coefficient of the horizontal data group, and the horizontal coefficients of the m horizontal data groups are summed and averaged to obtain the horizontal performance value HB; the variance of all elements in the vertical data group is calculated to obtain the vertical coefficient of the vertical data group, and the vertical coefficients of the n vertical data groups are summed and averaged to obtain the vertical performance value ZB; the variable coefficient BY of the detection area is obtained by the formula BY=v1×HB+v2×ZB, wherein v1 and v2 are both proportional coefficients, and v1>v2>1; the variable coefficient BY of the detection area is compared with the preset The variable coefficient BY is compared with the variable threshold BYmax: if the variable coefficient BY is less than the variable threshold BYmax, the corresponding detection area is marked as a normal area; if the variable coefficient BY is greater than or equal to the variable threshold BYmax, the corresponding detection area is marked as an abnormal area; the normal area and the abnormal area are sent to the data processing module; the gradient performance of the LCD screen is analyzed, and the variable coefficient is obtained after processing the horizontal data group and the vertical data group. According to the variable coefficient, the gradient display performance and the pure color display performance of the LCD screen are fed back, so that the detection area is differentiated by the variable coefficient, providing data support for the overall screen performance analysis.

[0036] The data processing unit is used to process and analyze the detection data of the liquid crystal display screen: mark the ratio of the number of abnormal areas in the detection object to the number of detection areas as the processing coefficient CL of the detection object, and compare the processing coefficient CL of the detection object with the preset processing thresholds CLmin and CLmax: if CL≤CLmin, it is determined that the gradient consistency of the processing object meets the requirements, and the corresponding processing object is marked as a qualified object; if CLmin<CL<CLmax, it is determined that the consistency of the processing object does not meet the requirements, the corresponding processing object is marked as a consistent optimization object, a consistent abnormal signal is generated and sent to the mobile phone terminal of the administrator; if CL≥CLmax, it is determined that the gradient of the processing object does not meet the requirements, the corresponding processing object is marked as a gradient optimization object, a gradient optimization signal is generated and sent to the mobile phone terminal of the administrator; the detection data of the liquid crystal display screen is processed and analyzed, and the overall performance of the liquid crystal display screen is fed back according to the proportion of the number of abnormal areas in the detection object, so as to sort and rework the detection objects with unqualified consistency and unqualified gradient, and improve the quality of the liquid crystal display screen.

[0037] Embodiment 2

[0038] like Figure 3 As shown, the data cleaning unit is communicatively connected to the data processing unit and the verification processing unit.

[0039] The data cleaning unit is used to clean the execution data of the machine vision group: generate a verification cycle, mark the camera of the machine vision group as a verification object, mark the process of performing a gradient display test on the verification object within the verification cycle as an execution process, and mark the execution process corresponding to the qualified object and the consistent optimization object as a verification process.

[0040] The verification processing unit is used to perform verification processing and analysis on the machine vision group: mark the number of times that the corresponding detection area of ​​the verification object during the verification process is marked as an abnormal area as an abnormal mark value, mark the number of times that the corresponding detection area of ​​the verification object during the verification process is marked as a normal area as a normal mark value, mark the ratio of the abnormal mark value to the normal mark value as an abnormal verification value, and compare the abnormal verification value with a preset abnormal verification threshold: if the abnormal verification value is less than the abnormal verification threshold, it is determined that the operating state of the verification object meets the requirements; if the abnormal verification value is greater than or equal to the abnormal verification threshold, it is determined that the operating state of the verification object does not meet the requirements, generate a visual verification signal and send the visual verification signal to the mobile phone terminal of the administrator; perform verification analysis on the machine vision group, regularly clean and process the verification data to obtain the abnormal verification value of each camera, and provide feedback on the necessity of camera verification based on the abnormal verification value to avoid a decrease in the accuracy of the detection result due to abnormal camera operation.

[0041] Embodiment 3

[0042] like Figure 4 As shown, a screen detection management method for a liquid crystal display production line based on machine vision includes the following steps:

[0043] Step 1: Perform a gradient display test analysis on the screen of the LCD screen: mark the screen of the LCD screen as a detection object, divide the detection object into several detection areas, generate a test image through software, display the test image in the detection area of ​​the detection object, and then randomly select a camera from the machine vision group to capture the image of the detection area, and perform data processing on the captured image to obtain m horizontal data groups and n vertical data groups;

[0044] Step 2: Analyze the gradient performance of the liquid crystal display screen and obtain the variable coefficient BY of the detection area, and mark the detection area as a normal area or an abnormal area through the variable coefficient BY;

[0045] Step 3: Process and analyze the test data of the LCD screen: mark the ratio of the number of abnormal areas in the test object to the number of test areas as the processing coefficient CL of the test object, and judge whether the gradual change and consistency of the process object meet the requirements through the processing coefficient CL;

[0046] Step 4: Clean the execution data of the machine vision group to obtain the verification process of the verification object;

[0047] Step 5: Perform verification processing and analysis on the machine vision group and obtain the abnormal verification value of the verification object, and determine whether the operating status of the verification object meets the requirements based on the abnormal verification value.

[0048] A screen inspection management system for a liquid crystal display production line based on machine vision, when working, marks the screen of the liquid crystal display as an inspection object, divides the inspection object into several inspection areas, generates a test image through software, displays the test image in the inspection area of ​​the inspection object, and then randomly selects a camera from a machine vision group to shoot an image of the inspection area, performs data processing on the shot image to obtain m horizontal data groups and n vertical data groups; analyzes the gradual change performance of the liquid crystal display and obtains a variable coefficient BY of the inspection area, and marks the inspection area as a normal area or an abnormal area through the variable coefficient BY; marks the ratio of the number of abnormal areas in the inspection object to the number of inspection areas as a processing coefficient CL of the inspection object, and determines whether the gradual change and consistency of the processing object meet the requirements through the processing coefficient CL; cleans the execution data of the machine vision group to obtain the verification process of the verification object; performs verification processing analysis on the machine vision group and obtains an abnormal verification value of the verification object, and determines whether the running state of the verification object meets the requirements through the abnormal verification value.

[0049] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

[0050] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0051] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A screen inspection and management system for a liquid crystal display production line based on machine vision, characterized in that: It includes a gradual change test unit, a gradual change analysis unit and a data processing unit; The gradient test unit is used to perform a gradient display test analysis on the screen of the liquid crystal display screen: mark the screen of the liquid crystal display screen as a detection object, divide the detection object into a plurality of detection areas, mark the horizontal pixel grids of the detection area as horizontal grids i, i=1, 2, ..., n, n is a positive integer, mark the vertical pixel grids of the detection area as vertical grids e, e=1, 2, ..., m, m is a positive integer, and generate a test image through software; display the test image in the detection area of ​​the detection object, and then randomly select a camera from the machine vision group to capture the image of the detection area, the machine vision group includes cameras set at different angles, and the captured images are processed to obtain m horizontal data groups and n vertical data groups, and the horizontal data groups and the vertical data groups are sent to the gradient analysis unit; The gradient analysis unit is used to analyze the gradient performance of the liquid crystal display screen and obtain a variable coefficient BY of the detection area, and mark the detection area as a normal area or an abnormal area through the variable coefficient BY; The data processing unit is used to process and analyze the detection data of the liquid crystal display screen; The pixel grids of the test image meet the following requirements: the pixel grids of the test image are distributed in m rows and n columns, the grayscale value of the first column of the pixel grid of the test image is zero, the grayscale value of the second column of the pixel grid is 0+255 / (n-1), the grayscale value of the second column of the pixel grid is 0+255 / (n-1)×2, and so on, and the grayscale value of the last column of the pixel grid is 255; The specific process of processing the captured image includes: enlarging the captured image into a pixel grid image and performing grayscale transformation, and then obtaining m horizontal data groups and n vertical data groups, and the process of obtaining the horizontal data group includes: selecting a row of pixel grids as a processing grid, marking the absolute value of the difference between the grayscale value of the second pixel grid and the grayscale value of the first pixel grid as the first processing value, marking the absolute value of the difference between the grayscale value of the third pixel grid and the grayscale value of the second pixel grid as the second processing value, and so on, marking the absolute value of the difference between the grayscale value of the nth pixel grid and the n-1th pixel grid as the n-1th processing value, and the n-1 processing values ​​constitute a horizontal data group; the process of obtaining the vertical data group includes: selecting a column of pixel grids as a marking grid, and the m marking grids constitute a vertical data group; The process of obtaining the variable coefficient BY of the detection area includes: performing variance calculation on all elements in the horizontal data group to obtain the horizontal coefficient of the horizontal data group, summing and averaging the horizontal coefficients of m horizontal data groups to obtain the horizontal performance value HB; performing variance calculation on all elements in the vertical data group to obtain the vertical coefficient of the vertical data group, summing and averaging the vertical coefficients of n vertical data groups to obtain the vertical performance value ZB; and obtaining the variable coefficient BY of the detection area by numerically calculating the horizontal performance value HB and the vertical performance value ZB; The calculation formula of the variable coefficient BY of the detection area is: BY=v1×HB+v2×ZB, where v1 and v2 are both proportional coefficients, and v1>v2>1.

2. According to the machine vision-based LCD production line screen detection management system of claim 1, it is characterized in that: The specific process of marking the detection area as a normal area or an abnormal area includes: comparing the variable coefficient BY of the detection area with the preset variable threshold BYmax: if the variable coefficient BY is less than the variable threshold BYmax, the corresponding detection area is marked as a normal area; if the variable coefficient BY is greater than or equal to the variable threshold BYmax, the corresponding detection area is marked as an abnormal area.

3. According to the machine vision-based LCD production line screen detection management system of claim 2, it is characterized in that: The specific process of the data processing unit processing and analyzing the detection data of the liquid crystal display screen includes: marking the ratio of the number of abnormal areas in the detection object to the number of detection areas as the processing coefficient CL of the detection object, and comparing the processing coefficient CL of the detection object with the preset processing thresholds CLmin and CLmax: If CL≤CLmin, it is determined that the gradient consistency of the processed object meets the requirements, and the corresponding processed object is marked as a qualified object; If CLmin<CL<CLmax, it is determined that the consistency of the processing object does not meet the requirements, the corresponding processing object is marked as a consistent optimization object, a consistent abnormal signal is generated and sent to the mobile terminal of the manager; If CL≥CLmax, it is determined that the gradual change of the processing object does not meet the requirements, the corresponding processing object is marked as a gradual change optimization object, a gradual change optimization signal is generated and sent to the mobile phone terminal of the manager.

Images