A non-sparse representation of the reference color based on image quality evaluation

The present invention discloses a non-reference color image quality evaluation method based on sparse representation, it belongs to the field of digital image processing technology. The present invention treats the evaluation image, in the HSV color space selected image block having the maximum correlation standard deviation of the pixel space, the image matching block atom dictionary, the dictionary every atom concentration maximum spatial correlation pixel having a trained FIG. standard deviation of the image block, DMOS value, a maximum standard deviation of three elements, and the method of matching pursuits to obtain the image to be evaluated in the sparse representation of the dictionary, the dictionary DMOS atom by atom to be selected and maximum standard the difference to build a color image quality evaluation. Atoms dictionary construction method of the present invention is simple, the extracted feature is more suitable for the human visual perception, the algorithm is simpler, more precise quality assessment.