LCD SURFACE DEFECT INSPECTION USING MACHINE VISION

To promote product quality of TFT-LCD in mass manufacturing, development of an automatic defect inspection system to replace human inspection is required. In this paper, we use machine vision for automatic inspection and present several research results for TFT-LCD defects inspection and classification. For MURA defects inspection, Jiang et al. (2003) proposed to use ANOVA and EWMA control charts for detecting non-uniformity areas in a panel, and to design a weighted index mask matrix for MURA defects classification. For Micro defects, including pinholes, scratches, particles and fingerprints on the surface of TFT panels, Lu and Tasi (2004) proposed a global approach that uses an SVD-based image reconstruction technique. In the experiment, we have evaluated a variety defects including MURA, pinholes, scratches, particles and fingerprints on TFT panel surfaces, and the result reveals that the proposed methods are effective for LCD defects inspection.