Objective analysis method for just noticeable change step length of stereo images

Abstract

The invention discloses an objective analysis method for just noticeable change step length of stereo images. The objective analysis method comprises the following steps of: training each distorted stereo image in a distorted stereo image set by using different weight proportion combinations to obtain a support vector regression training model of the different weight proportion combinations; testing any test image by utilizing the support vector regression model obtained by training; in a condition that a left-view image has a constant and invariable quality, measuring a critical quality point of a right-view point image when the quality variation of the stereo image is sensed by human eyes so that a maximum variation range in which the quality of the right-view image can be decreased relative to the quality of the left-view image can be determined when the stereo image is coded. The objective analysis method not only can achieve the purpose of improving coding efficiency by decreasing quality of the right-view image, but also ensures that an observer cannot sense the reduction in the quality of the right-view image through a stereoscopic visual masking effect, thereby ensuring the integral quality of the stereo image.

Description

technical field [0001] The invention relates to a method for measuring and analyzing the visual perception characteristics of a stereoscopic image, in particular to an objective analysis method for the smallest perceivable change step of the stereoscopic image. Background technique [0002] Compared with two-dimensional (2D, Two Dimensional) video display, stereoscopic / three-dimensional (3D, Three Dimensional) video display is more compatible with human vision, and it makes people feel three-dimensional and immersive when watching images on the screen. In recent years, governments and enterprises in the United States, Europe, Japan, and South Korea have invested heavily in the research and development of three-dimensional television / three-dimensional television (3DTV, Three Dimensional Television), including the National Natural Science Foundation of the United States, the U.S. Department of Naval Research and Energy, The American Advanced Visual Technology Center, the 3DTV ...

AI Technical Summary

Problems solved by technology

[0004] At present, the minimum perceivable change step size when the human eye can perceive the change of stereo vision is mainly determined through subjective experiments. However, subjective experiments are easily affected by various factors such as objective conditions, subjective emotions, and observers' own conditions, and the evaluation results are not stable enough. Subjective experiments are time consuming and costly

Images