Innovative Color Interpolation Using Fuzzy Logic and Linear Regression

To reduce the cost of digital cameras, the color filter array (CFA) is usually coated upon a single-chip image sensor. Each pixel of the image sensor can sense only one of the R, G and B colors under CFA. The two missing colors of a pixel are estimated using the CFA interpolation algorithm. The interpolation algorithm may generate color distortion and degrade the quality of interpolated images. In this paper, we propose a new algorithm “fuzzy logic and linear regression method (FLLRM)” to solve this problem.

We use fuzzy logic method based on the effect between pixels to avoid color distortion around the edges. The linear regression method is used to process the interaction effect among R, G and B channels. The proposed method can successfully eliminate false color, color moire, aliasing and color shifts in the full color camera output. Finally, we implement our method on the digital camera platform that is designed by Industrial Technology Research Institute (ITRI). The processed image shows a very good quality, which also demonstrates our approach is a better CFA interpolation algorithm than the previous processed methods.