摘要

Through analyzing three dimensional characteristics of stereo images, such as human vision and binocular difference information, a new stereo image quality evaluation method is put forward, which combines stereo vision characteristics and structural similarity of images. This method considers the influence of the following factors on the stereo image: brightness, texture details and dimensional position. The weighted mean structure similarity and statistics method are combined to work out an initial value, which is revised with binocular difference containing depth information of scenery, and finally an evaluation index of image quality is got. Experimental results show that this method is in accordance with human visual characteristics, and consistent with the result of subjective evaluation preferably.

全文