Color-based structural similarity image quality assessment

It is known that the number of colors that can be distinguished by human eye is much more than that of the grayscale levels and more information is contained in color image than what is contained in grayscale one. By studying the Structural Similarity Index (SSIM), it is found that it fails in estimating the quality of color images with color-related noises. Based on this, we develop an improved method which is called Color-based Structural Similarity. The proposed method was validated on the TID2013 image database and the experimental results show that the quality scores given by the proposed method are more consistent with HVS than the classical one.