Color Image Quality Assessment Based on Quaternion Representation for the Local Variance Distribution of RGB Channels

In this paper, Quaternions are used to describe the structural information of a color image in order to assess its quality. The representation method is different from conventional models in that some information that is sensitive to human eyes is reinforced by using local variance. It encodes the local variance distribution of the RGB channels of a color image into the three imaginary parts of a quaternion. The luminance layer of the color image is taken as the real part of the quaternion. The distance between the singular value feature vectors of the source image block and the distorted image block which are described by quaternion matrices is calculated. The final assessment result is calculated by using the mid point of the distances. The experimental results show that the assessment results of the proposed assessment method are more consistent with the Human Visual System than those of the conventional assessment methods.