An intelligent defect inspection technique for color filter

Abstract

Automatic defect inspection systems are becoming more and more important in industrial production lines. Especially in electronics industry, an attempt is often made to achieve almost 100% quality control of all components and final goods. Here we are interested in the defect inspection of color filter, which is one of components in TFT-LCD module and gives each pixel of LCD its own color. The difficulties in the defect inspection of color filter are its complex texture and demand for high-speed processing. In this paper, we propose a neural-fuzzy-inference-network (NFIN)-based defect inspection algorithm to detect the materials with regular pattern such as color filter. The NFIN, which is basically a fuzzy inference system and its fuzzy rules and corresponding parameters can be learned by neural network automatically, is a good alternative to achieve the defect inspection. Experimental results show that the proposed algorithm is a promising method to detect the defects of color filter. The proposed algorithm can apply to not only the detection of color filter but also the detection of web materials.