Image quality assessment based on nonsubsampled contourlet transform

Objective image quality assessment (QA), which automatically evaluates the image quality consistently with human perception, is essentially important for numerous image and video processing applications. In this paper, based on the characteristics of nonsubsampled contourlet coefficients of images and the correlativity indexes, a novel image quality assessment is proposed. Firstly, the reference image and the distorted images are decomposed into several levels by means of nonsubsampled contourlet transform respectively. The nonsubsampled contourlet coefficients of the reference image (the distorted images) are as the reference sequences (the comparative sequences). Secondly, calculate the correlativity indexes between the reference sequences and the comparative sequences respectively. Moreover, image quality assessment vector of every distorted image can be constructed based on the correlativity indexes and image quality can be assessed. The algorithm makes full use of perfect integral comparison mechanism of the correlativity indexes and the well matching of nonsubsampled contourlet transform with multi-channel model of human visual system. Experimental results show that the proposed method improves accuracy and robustness of image quality prediction.