Image-Based Inspection System for Detection of Glass Substrate’s Edge Defects
Large size TFT-LCD (Thin-Film Transistor Liquid-Crystal Display, which is commonly called as panel) is made after lots of processes and is cut based on the required size in the last process. Deckle-edge (Burr) and other defects were produced by cutting. Therefore, it needs to examine the products’ defects before grinding the edges or the following processes. This research developed a new type and high-efficiency apparatus for panel-edge inspection. The defects are sorted based on three types “Fractures, Cracks and Dark-spots” by using image-subtraction, binary thresholding, and blob-analysis methods. The proposed system can improve the efficiency on panel-edge inspection for panel manufacturers. The production of flat panel-edge inspection can be increased to over 40% which can resolve the bottleneck processes for panel industry.
Dongye Sun, Wen-Pei Sung and Ran Chen
Q. C. Hsu and C. M. Lin, "Image-Based Inspection System for Detection of Glass Substrate’s Edge Defects", Applied Mechanics and Materials, Vols. 121-126, pp. 3970-3976, 2012