Artifact reduction of interpolated color filter array images using modified mean-removed classified vector quantization

To reduce the artifacts due to color filter array (CFA) interpolation, a modified mean-removed classified vector quantization algorithm is proposed. The algorithm extends and modifies vector quantization to discover the relationships between the G channels of interpolated images and their corresponding original versions using the information from CFA images. The discovered relationships are stored in two codebooks and are used to improve the edge and texture quality of G channels of interpolated images. After the interpolated G values are refined, the interpolated R and B values can also be improved using the refined G values. The experimental results show that the proposed algorithm can effectively reduce the artifacts of interpolated CFA images. Comparing our method and the best CFA interpolation algorithm as far as we know, the PSNR improvements of R, G, B channels are 0.89 dB, 0.71 dB, and 0.74 dB, respectively.