A Selective Interpolation Scheme for Mobile Camera Sensors

Abstract

Pictures can be taken anywhere with cameras built in mobile phones, but it is not easy to make a good quality of picture with the built-in camera, especially when enough light is not provided. One of the reasons for the degradation is that with little light, noise increases and the interpolation for Bayer patterned image does not produce good quality of image. In this paper we compared the performance of interpolation algorithms for Bayer patterned image sensors under the various lighting conditions and our experimentation showed that performance of interpolation algorithms depend on the lighting conditions heavily. Based on this observation, we proposed an interpolation scheme that selects the best interpolation algorithms according to the input lighting condition. This scheme can be easily implemented and embedded in the signal processor to improve the quality of pictures for mobile cameras.