Large-size liquid crystal display screen defect detecting method based on machine vision

Abstract

The invention discloses a large-size LCD (liquid crystal display) screen defect detecting method based on machine vision. An LCD screen is zoned, one camera is arranged in each zone and calibrated toacquire a distortion factor matrix, and a homography matrix is obtained; a pure color picture is displayed on the LCD screen, each camera shoots an LCD screen image in each corresponding zone, distortion correction is performed with the adoption of the distortion factor matrix, all LCD screen images are transformed in the same coordinate system according to the homography matrix and spliced to obtain a complete LCD screen image, the edges of the LCD screen are detected, a regional area of the LCD screen is extracted, and defect detection is performed after background texture inhibition and image intensification are performed. Defect detection of the large-size LCD screen can be effectively realized, and better universality is realized.

Description

Technical field [0001] The invention belongs to the technical field of liquid crystal screen defect detection, and more specifically, relates to a large-size liquid crystal screen defect detection method based on machine vision. Background technique [0002] The LCD screen is currently a very commonly used display device. Before the LCD screen leaves the factory, the LCD screen needs to be tested for defects. At present, most relevant LCD screen manufacturers rely on traditional manual inspection methods to detect LCD screen defects. The manual inspection method is mainly based on the observation and feeling of the human eye on the LCD screen and relying on the experience of the inspector to determine and quantify the defects of an LCD screen. The disadvantages of this method are very obvious. It is highly subjective and susceptible to interference from the external environment, so there is no uniform judgment standard for the quantification of defects, that is, the manual dete...

AI Technical Summary

Problems solved by technology

The manual inspection method is mainly based on the observation and feeling of the LCD screen by the human eye and the judgment and quantification of the defects of an LCD screen based on the experience of the inspector. The disadvantages of this method are very obvious. With a lot of subjectivity and susceptible to interference from the external environment, there is no unified judgment standard for defect quantification, that is, the manual detection method lacks objectivity

The second is that if the inspector works for a long time, he will be prone to fatigue, that is, he needs to rest for adjustment after working for a period of time, which will reduce the detection efficiency and increase the production cost

[0003] As consumers continue to pursue high-quality display effects and visual experience, the current LCD display industry is developing in the direction of high-definition, large-size, and wide-viewing angles, which further increases the difficulty of defect detection for large-size LCD screens

Images