An improved method of SSIM based on visual regions of interest

Image quality assessment is always a hot research topic in the field of image processing. Structural Similarity Image Measurement (SSIM) is an image quality assessment algorithm with the advantages of simplicity, high efficiency and better consistence. Its evaluation of performance is better than PNSR and MSE. However, it often fails when assessing badly distorted or cross distorted images. In this paper, we proposed a new method on the improved method of SSIM and the method of based on visual region of interest combination. This improved method of SSIM takes the histogram concentration as the main structural information of an image. It used histogram concentration to calculate the fuzzy degree of the image. Finally, we can obtain the structure similarity value of the image. The experiment results show that, compared with the SSIM model, the proposed RoiHSSIM model is more close to the human visual system and can access the quality of fault images more precisely.