A Method for Positioning Mark Point on Liquid Crystal Glass Based on Machine Vision
Fei Zou, Xiaodong Liu, Jun Gong · 2020
The positioning of liquid crystal glass is very important in the production process of liquid crystal glass. In order to improve the assembly efficiency of liquid crystal backlight screen, this paper proposes a liquid crystal glass positioning algorithm based on machine vision to realize fast and accurate positioning of Mark points (cross shape) on liquid crystal glasses. The algorithm first solves the problem of image fogging and uneven illumination by using the single-scale Retinex algorithm for the original image and the template image, and then determines the Region of Interest (ROI) based on gray-scale template matching. Next, perform Mean filtering, threshold segmentation and the open operation on the ROI so that the independent and complete Mark point is obtained. Finally the minimum external rotation rectangle fitting algorithm is used to fit and locate the Mark point. The experiment results show that the method in the paper can quickly and accurately locate the Mark point on the liquid crystal glass. In the sample experiment, the recognition success rate reached 98.04%, the positioning success rate reached 94.12%, and the positioning error of angle and the Mark point center point coordinate was controlled within 1 pixel unit and 0.3° respectively.