Region Mura Detection using Efficient High Pass Filtering based on Fast Average Operation

Abstract A Mura (also called “alluck”) defect that causes local deterioration of colors or brightness on a LCD(liquid crystal display) panel is one of the critical issues in display manufacturing. In this paper, we propose mura detection method using frequency filtering. As a mean of frequency filtering to eliminate the low frequency background term, subtraction of the averaged image from the raw one is utilized. And a fast averaging algorithm has been devised for reducing computational costs. The validity and the performance of the proposed method is discussed through experimental studies on actual LCD panel mura defects.