Objective image quality assessment based on image content contrast perception
- Received Date:
04 March 2020
Abstract: Image quality assessment (IQA) plays a much important role in acquisition, storage, transmission and processing of image and video. In order to improve the IQA efficiency and achieve better consistency between the subjective and objective IQA results, an IQA method based on image content contrast perception is proposed based on the characteristics of human visual perception and the gray and gradient, local contrast, and blur of image. In this method, firstly, combining with the characteristics of human visual perception, based on the definition of contrast in physics, a definition for image quality is proposed. Then, based on the gray gradient co-occurrence matrix, a novel concept, that is the entropy of the gray gradient of image, and its calculation method are proposed; and based on the gray-gradient entropy, local contrast and blur of the image, a method for describing the image content and its visual perception is proposed. Finally, based on the image content and the definition of image quality, an IQA method and its IQA mathematical model are proposed. In addition, 119 reference images and 6395 distorted images from CSIQ, LIVE, TID2008, TID2013 and IVC5 image databases are used for test. Meanwhile, the influence of 52 distortion types on IQA is analyzed. The experimental results show that the accuracy PLCC of the proposed IQA model can achieve 0.8616 at least and 0.9622 at most in the five databases. At the same time, in order to illustrate the advantages of the proposed IQA model, it is compared with seven existing typical IQA models in terms of accuracy, complexity and generalization performance, which show that the comprehensive benefit of the proposed model is better than those of seven existing IQA models. Experimental tests and comparison with the existing IQA models show that the proposed IQA method is effective and feasible, and the proposed IQA model is an excellent IQA model.