Blind reference tone mapping image quality evaluation method

Abstract

The invention discloses a blind reference tone mapping image objective quality evaluation method. According to the method, first, a high-brightness region and a low-brightness region of a tone mappingimage are extracted, and a detail information amount of the local regions and a detail information amount of a global region are combined to measure detail information distortion caused by tone mapping; second, natural scene statistical characteristics of a brightness channel and a yellow channel of the tone mapping image are used to measure naturalness distortion of the brightness and color of the image respectively; third, brightness and color characteristics of the tone mapping image are extracted from the aesthetic perspective; and last, a random forest is used to perform regression on all the characteristics, and image quality is predicted. Therefore, objective quality evaluation of the tone mapping image for blind reference is realized, the evaluation effect is obviously improved, and the image has better consistency with human eye visual perception.

Description

technical field [0001] The invention relates to the technical field of image quality evaluation, in particular to a blind reference tone mapping image quality evaluation method. Background technique [0002] Generally speaking, the dynamic range of luminance that the human visual system can accept in the same scene can reach 6 orders of magnitude, while the dynamic range that can be represented by traditional low dynamic range images does not exceed 3 orders of magnitude, which cannot present the natural scene to the human eye. Realistic feelings, for which people have pushed the development of high dynamic range images. High dynamic range images have a wider dynamic range than low dynamic range images, and their levels of expression are richer. The details of high-brightness areas and low-brightness areas in the scene can be well preserved, but dedicated high dynamic range display devices The development process is complicated, the cost is high, and it is difficult to popu...

AI Technical Summary

Problems solved by technology

Therefore, the direct use of traditional blind reference methods cannot accurately measure the degree of image distortion.

Based on this, Gu et al. proposed the BTMQI algorithm from three aspects of the information content, naturalness and structure of the tone-mapped image. This method defines the characteristics that the tone-mapped image should have based on the characteristics of the high dynamic range image. Nice effect, but it ignores color distortion

[0006] Based on the above analysis, the blind reference tone mapping image quality evaluation method has more research value, but the existing methods are too few and incomplete, and their subjective consistency needs to be improved

Images