Image Quality Assessment Based On Gradient Similarity

The main contents of this paper is the image quality assessment algorithm. The image quality assessment algorithm is a process which compares the differences between the reference and the distortion image, and finally to get the degradation degree of distortion image. This is the two main parts of the contents in this paper.In the modern information era, the significance of the information in our work and life is prominent. While the information as a form usually depends on some certain carrier. And we usually begin the research by the carrier of the information to make it carried out smoothly. As we know the images as one of the forms of the information is inevitable to be the important object.While digital images are usually affected by variety of distortions during acquisition, transmission, compression, restoration or enhancement, etc. Therefore, image quality assessment (IQA) is useful in many applications. How to accurately judge the image quality and how to judge the performance of the algorithm in the image technology has become one of the core problems in the image processing technology.To that end, we launched a study in this paper. In our study, the new algorithm through extracting the main representative information of the image which is including the gradient and luminance information, and then we get the integration assessment by comparing the origin reference image. Finally, we need fit the object and the subject algorithm to confirm that our new IQA algorithm is in accord with the human visual system.The main innovation of this paper in comparison to the existing works are as follows:1) We demonstrate that gradient information that captures both contrast and structure of the image;2) our scheme matches better with the contrast masking, particularly for the cases when both masked and masking signals are small; and3) we devise an adaptive approach to integrate the different components of distortion.