The Defect Detection Using Human Visual System and Wavelet Transform in TFT-LCD Image

The thin film transistor liquid crystal display (TFT- LCD) image has non-uniform brightness, which is the major difficulty in finding the defect region called Mura. To facilitate Mura segmentation, globally widely varying background signal has to be flattened and Mura signal must be enlarged. In this paper, Mura signal magnification and background signal flattening method is proposed using wavelet coefficients processing and the properties of human visual system (HVS). The wavelet approximation coefficients are used for background signal flattening while wavelet detail coefficients are employed to magnify Mura signal based on adapted contrast sensitivity function (ACSF). For the enhanced image, tri-modal thresholding segmentation technique is used for finding dark and white Mura at the same time. For final reliable defect confirmation, false region elimination algorithms based on Weber´s law are also proposed. By the experimental results of TFT-LCD image, the proposed algorithms can have promising results and can be applied to the real automated TFT-LCD inspection system.