Color Signal Estimation of Surrounding Scenes from Human Corneal Surface Images

Abstract

A method is proposed for estimation of the wide-area color signals of the surrounding scene in the image reflected on the corneal surface. A digital color camera and the human eyeball are used to develop a simple imaging system without the need for any specialized devices. First, the authors introduce the corneal imaging system. Second, they correct the geometric distortion of the reflected image using gnomonic projection. Third, the color signals in the surrounding scene are estimated from the RGB image data according to the Wiener estimator. The corneal image consists of a combination of specular and diffuse reflections from the cornea and the iris. The desired color signals are recovered only from the specular reflection component after removing the diffuse reflection component. Finally, experiments are performed in several different conditions. The results demonstrate that the proposed corneal imaging method can estimate reliable color signals over a wide area.