Image reconstruction for color contact image sensor (CIS)

Abstract

The color contact image sensor is often used to capture the surface of some materials for the defect detection in industry. However, the special imaging mode leads a special image pattern of the color contact image sensor. This pattern of the sensor can be used to increase the resolution of the image, while none of the algorithms is able to properly process it, recently. This paper presents an approach for the reconstruction of the color contact image sensor. We combine the sparse prior that often used in super-resolution and the inter-channel correlation prior that the majority of image demosaicing algorithms used to solve this problem. Extensive experiments on simulated image and the real image captured by color contact image sensor show that our method achieves good results in terms of both objective and human visual evaluations.