No-reference remote sensing image quality assessment based on the region of interest and structural similarity

Blur and noise are two common distortion factors which affect remote sensing image quality. And make it difficult to assess the remote sensing image quality. The Structure Similarity(SSIM) algorithm is simple, high efficient and accurate. However, it does not work well when there is cross distortion in the image. To deal with the problem of SSIM algorithm treating different regions of image identically, this paper considered the perceptual characteristics to different content and masking effect. The proposed method is the perceptual weighting used in the region of interest and based on SSIM algorithm. The experiment shows that, compared with the Peak Signal-Noise Rate(PSNR) index, the proposed index has good consistence with the Structure Similarity(SSIM) index, and can make an effective and correct evaluation of image with both noise and blur. This is an accurate and reliable no-reference remote sensing image quality assessment mothed, which is easy to implement.