摘要：With the increase in the usage of LCD’s. FPD (flat panel display) are becoming widely acceptable and popular as display devices growing at a rapid pace and making it a promising investment opportunity. In this paper, we propose a machine vision approach for automatic inspection of Mura defects. The proposed method is based on wavelet based multiresolution structure and an efficient image restoration method for inspection of local defects present in LCD surfaces. The proposed method consist of selecting the low frequency and band frequency components or the combination of band frequencies at different resolution levels and giving weights to each band to enhance the threshold level for defect extraction.
