Gamma curve determination for each individual pixel on high-resolution flat panel display

Abstract

This paper presents an automated system to determine gamma curves of each individual pixel on a highresolution flat panel display. The system consists of a panel and its aligned area scan camera located at a fixed position. To localize a panel pixel position in an image captured from the aligned camera, a mapping function estimation scheme is proposed between the panel and the camera coordinates. The mapping function is modeled by combining both 2-D perspective transform and lens distortion and is estimated via feature-based registration. Intensity values of individual panel pixel are then measured for sampled gray-levels. To alleviate intensity interferences from neighboring pixels, we propose to use evenly spaced dot images as panel input. By using the measured intensity values of each panel pixel, gamma curves are determined to examine the characteristics of defects on the panel, if any. Experimental results show the relevance of the proposed method.