Evaluation de la qualité d’image

Abstract

Image quality assessment (IQA) plays a major role in a broad range of applications. Evaluating the adequacy of a quality image for a given application (image or video processing applications) is a requisite. However, objective quality assessment is far from being a solved problem. In this work, we have addressed the problem of assessment of images quality. An approach based on edge-region information, distorted and displaced pixels have been developed for image quality assessment. First, the distorted and original images are divided into blocks of 11×11 pixels, andthen we compute distorted pixels and displacement which can be used to compute the global error. Simulation results with well known methods show that the performance of the proposed technique is found to be better than previously metrics which do not use the pixel displacement in calculation. Thus, the results indicate that this global method can be helpful in the assessment and comparison of quality of images. In the conducted experiments, the proposed IQA method has showed the aptitude to evaluate the image quality according towhat a human observer would do.