No Reference Evaluation of Image Quality Based on Structure Self-Similarity and Wavelet Transform Coefficients

Image quality evaluation is an important topic in image processing. In the paper, a novel no reference image quality evaluation algorithm is introduced, which is based on wavelet transform (WT), information entropy and structure self-similarity (SSS). The algorithm composes of two steps. The first step is separating image into 4^N blocks equally based on the self-similarity And then, wavelet transform coefficients in different layers were used to calculate the information entropy, which is the evaluation criterion of image quality. Experiment results show the excellent effect of the algorithm without any reference image.