A Novel Defect Detection Method of Liquid Crystal Display Based on Machine Vision

Abstract

As an important information display tool closely related to people's daily life, the liquid crystal display (LCD) has become an inseparable part of people's lives. In the manufacturing process of LCD, screen defect detection is an indispensable step which directly affects the yield and quality of LCD. In order to improve the accuracy and efficiency of defect detection for LCD screen, this paper proposes a novel defect detection method for LCD based on machine vision. Firstly, preprocessing operations including grayscale, binarization, filtering and dilation are used to reduce background noise and enhance the useful features of LCD screens. Secondly, the maximum connected region (MCR) and minimum external rectangle (MER) are adopted to initially locate the position of the LCD screen; Then, the affine transformation is introduced to correct the tilted screen and horizontal projection (HP) and vertical projection (VP) are presented to extract the LCD screen. Finally, a regional template matching algorithm is proposed to detect defects of LCD screens. Experiments show the effectiveness and robustness of the proposed method.