CN-115222728-B - Screen detection method and device based on vision technology and computer equipment

CN115222728BCN 115222728 BCN115222728 BCN 115222728BCN-115222728-B

Abstract

The application relates to a screen detection method and device based on a vision technology and computer equipment. In one embodiment, a global image of a screen is obtained through a global camera, position information of bad points is obtained according to the global image, then a re-judging camera is moved to a designated position, the focal length of a lens of the re-judging camera is adjusted for multiple times, then a re-judging photo set with different depths is obtained through shooting, the re-judging photo set is analyzed, and the height information of the bad points is compared with the height information of the screen for judgment. In this way, in the scheme, a complex judgment camera is used for focusing for multiple times to obtain multiple complex judgment photos with different depths, and the height information of the bad points can be directly identified by a software algorithm. The height of the bad points is combined with screen information to accurately judge whether the detected liquid crystal screen is qualified or not, so that the detection accuracy is improved. In addition, the process can be completely realized through a software algorithm, so that the manual rechecking process is reduced, the labor cost is saved, the detection efficiency is greatly improved, and the detection time is reduced.

Inventors

Ding Xiaoshuai
XIONG XING
WU HAOCHEN

Assignees

苏州华兴源创科技股份有限公司

Dates

Publication Date: 20260508
Application Date: 20220819

Claims (10)

1. A screen detection method based on vision technology, comprising: Acquiring a global image of a screen through a global camera; Acquiring position information of bad points according to the global image; Moving a re-judging camera to a designated position according to the position information, wherein the re-judging camera uses a liquid lens; Adjusting the focal length of the liquid lens, and photographing the bad points after adjusting the focal length each time to obtain a complex judgment photo set; judging that a picture with the clearest focusing pixel layer in the complex judgment picture set is taken as a picture of the pixel layer according to the definition and gray level of the bad point in the complex judgment picture set, taking the picture with the clearest focusing pixel layer in the complex judgment picture set as a picture of the bad point, and calculating to obtain the height information of the bad point according to the interval relation between the picture of the bad point and the picture of the pixel layer in the complex judgment picture set and a preset focusing height difference; And acquiring screen height information, and judging whether the screen is qualified or not according to the screen height information and the height information of the bad points.
2. The screen detection method according to claim 1, wherein the acquiring the position information of the defective point from the global image includes: And setting a screen display area to be white through binarization processing, taking the screen display area to be black externally, and obtaining an enlarged graph after independently taking out the white area, and obtaining the position information of the bad points according to the enlarged graph.
3. The screen detection method according to claim 1, wherein moving the re-judging camera to the specified position comprises: converting the position information into pixel coordinates by using a software algorithm; calibrating the image coordinates of the global camera and the initial coordinates of the complex judgment camera according to a nine-point calibration method; Obtaining an actual coordinate corresponding to the pixel coordinate; and moving the complex judgment camera to the actual coordinates.
4. The screen detection method according to claim 1, wherein the adjusting the focal length of the liquid lens includes: And adjusting the input voltage of the liquid lens, changing the internal curvature of the liquid lens, enabling the height difference of each focusing to be a fixed value, and taking a picture of the defective point after each adjustment to obtain the complex judgment picture set.
5. The screen detection method according to claim 4, wherein the calculating the height information of the defective point relative to the screen pixel layer according to the interval relation between the defective point photo and the pixel layer photo in the complex judgment photo set and the preset focusing height difference includes: Taking the picture of the pixel layer as a picture taking a position of a pixel datum plane at 0 micrometer as a focus, taking the next repeated picture as a picture taking an X height, and so on, wherein X is the focusing height difference; And calculating the height difference between the defective point photo and the pixel layer photo according to the quantity relation between the defective point photo and the pixel layer photo interval, and determining the height information of the defective point.
6. The screen detection method according to claim 1, wherein the determining whether the screen is acceptable comprises: comparing the height information of the bad points with the screen height information; If the defective point is positioned above the screen, judging that the screen is qualified, and if the defective point is positioned below the screen, judging that the defective point is an internal foreign matter, and judging that the screen is unqualified.
7. The screen detection method of claim 1, wherein the screen comprises a cell phone screen, a tablet screen, a television screen, a computer screen.
8. A screen detection device based on vision technology, comprising: the image acquisition module comprises a first image acquisition unit and a second image acquisition unit; the first image acquisition unit is used for acquiring a global image of the screen, and the second image acquisition unit is used for acquiring a re-judging photo of a bad point on the screen; the control module is used for controlling the movement of a jig carrying a screen and controlling the movement of the first image acquisition unit and the second image acquisition unit; the image processing module is used for processing the global image information, calculating the position coordinates of the bad points, carrying out algorithm identification on the definition of the complex judgment photo, judging that the photo with the clearest focusing pixel layer in the complex judgment photo set is taken as the pixel layer photo, taking the photo with the clearest bad point as the bad point photo, and calculating to obtain the height information of the bad points according to the interval relation between the bad point photo and the pixel layer photo in the complex judgment photo set and the preset focusing height difference; And the display module is used for displaying the information output by the image processing module.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

Description

Screen detection method and device based on vision technology and computer equipment Technical Field The disclosure relates to the field of screen detection, and in particular relates to a screen detection method, device and computer equipment based on vision technology. Background With the progress of liquid crystal technology, the cost of liquid crystal panels is continuously reduced, and more electronic devices begin to adopt liquid crystal screens. Accordingly, the increase in the amount of shipping makes it a difficult task to quickly and efficiently qualify a screen. Currently, the main liquid crystal screens are an LCD screen and an OLED screen, wherein the OLED screen comprises a pixel layer, a polaroid and surface glass, and the LCD screen is provided with a backlight and a polaroid below the pixel layer. Many defects are found in the process of optically and automatically inspecting the lcd screen AOI (automatically optical inspection), and the areas of the defects are generally smaller than one or more pixels according to the definition of each manufacturer. In the conventional lcd screen detection, the illuminated screen is photographed by a 2D area camera, and possible bad points in the photographed picture are analyzed, but since the area camera imaging is 2D plane imaging, the bad points can only be represented in the form of small points. However, the reason for causing the defective spot may be a defect from the pixel layer, or may be that dust or foreign matter is adhered above the surface glass, the defect of the pixel layer represents a problem of liquid crystal quality, and the dust or foreign matter on the surface glass can be erased without affecting the screen quality, so that the screen with the defective spot in the traditional detection process needs to enter a manual re-judgment process to manually confirm whether the defective spot causes the product to be unqualified, thus the detection process is more complicated, and meanwhile, the risk of missed detection exists. Therefore, a detection method capable of accurately detecting the cause of the defective spot at one time is needed. Disclosure of Invention The disclosure provides a screen detection method, a device and computer equipment based on a visual technology, so as to solve the problem that the existing detection method is difficult to distinguish reasons of poor points of a screen. The technical scheme of the present disclosure is as follows: according to a first aspect of embodiments of the present disclosure, there is provided a screen detection method based on vision technology, including: Acquiring a global image of a screen through a global camera; Acquiring position information of bad points according to the global image; Moving a re-judging camera to a designated position according to the position information, wherein the re-judging camera uses a liquid lens; Adjusting the focal length of the liquid lens, and photographing the bad points after adjusting the focal length each time to obtain a complex judgment photo set; Obtaining the height information of the bad points according to the complex judgment photo set; And acquiring screen height information, and judging whether the screen is qualified or not according to the screen height information and the height information of the bad points. In one embodiment, the acquiring the location information of the bad point according to the global image includes: And setting a screen display area to be white through binarization processing, taking the screen display area to be black externally, and obtaining an enlarged graph after independently taking out the white area, and obtaining the position information of the bad points according to the enlarged graph. In one embodiment, the moving the complex camera to the designated location includes: converting the position information into pixel coordinates by using a software algorithm; Calibrating the image coordinates of the global camera and the initial coordinates of the complex judgment camera according to a nine-point calibration method; Obtaining an actual coordinate corresponding to the pixel coordinate; and moving the complex judgment camera to the actual coordinates. In one embodiment, the adjusting the focal length of the liquid lens includes: And adjusting the input voltage of the liquid lens, changing the internal curvature of the liquid lens, enabling the height difference of each focusing to be a fixed value, and taking a picture of a bad point after each adjustment to obtain a complex judgment picture set. In one embodiment, the obtaining the height information of the bad points according to the complex photograph set includes: judging a picture which is most clear in focusing of a pixel layer in the complex judgment picture set as a picture of the pixel layer according to the definition and the gray level of the bad point; taking the picture of the pixel layer as a picture taking a position of a pixel reference