Brightness and color correction for dual camera image registration

Asymmetric dual camera system needs a preprocessing for registration in two input images due to the difference of brightness and color information. To solve this problem, this paper presents a framework that corrects the brightness and color between two input images. The proposed framework consists of three steps : i) selection of the reference and source images based on the contrast, ii) brightness enhancement using the brightness compensation map to minimize the difference of average brightness level of two input images, and iii) color correction of source image using color information of the reference image to compensate the disparity based on the illumination condition and exposure parameters. Experimental results show that proposed brightness and color correction method improves the performance of image registration in an asymmetric dual camera system.