Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : June 06, 2021 - June 08, 2021
Coding illumination-independent color information is an important feature for robotic vision because the illumination condition strongly affects the apparent color of objects. In this study, we developed an image sensor system with color constancy based on the center / surround retinex model. The system uses the Michaelis-Menten equation in place of the logarithmic function used in the original model in order to encode the intensity information efficiently. The system is composed of a CMOS image sensor and a field-programmable gate array. To evaluate the system output, we presented colored objects to the system under various lighting conditions, and examined the change in image regions with a particular color defined by the system output. The comparison between the change in regions computed by the conventional method and that computed by the proposed system suggested that the performance of the proposed system is better in terms of color constancy.
