A novel contourlet-based No-Reference Image Quality Assessment metric

摘要

No-Reference (NR) Image Quality Assessment (IQA) is of fundamental importance to numerous image processing applications. Whose goal is to evaluate the image quality without a reference image and without knowing the distortion present in the image. To solve this problem, we extract a set of statistical features from a computed image contourlet representation to learn a new no-reference image quality assessment model. In addition, we attach more attention on the distortion of image salient regions considering characteristic of human visual system. Our results indicate that these features are sensitive to the presence and severity of image distortion. Operating within a 2-stage framework of distortion classification followed by quality assessment, a distortion classification and quality prediction model is trained by a Support Vector Machine (SVM). The result of Experiment on LIVE database shows that the statistical performance of our algorithm is comparable to the state-of-the-art NR/FR IQA algorithms.