Tone mapping image quality objective evaluation method based on global characteristic

Abstract

The invention discloses a tone mapping image quality objective evaluation method based on a global characteristic, wherein the method comprises two processes of a training period and a testing period. Influences of a natural scene statistics characteristic and a color statistics characteristic to tone mapping are considered. A global characteristic vector of a tone mapping image is extracted. Then support vector regression is utilized for training the global characteristic vectors of all tone mapping images in a training image set, thereby constructing a quality predication model. In the testing period, through calculating the global characteristic vector of the tone mapping image for testing, and a quality objective evaluation predicated value of the tone mapping image is obtained through predicating according to the quality predication model which is constructed in the training period. Because the obtained global characteristic vector information has relatively high stability and furthermore can well reflect the quality change condition of the tone mapping image, correlation between an objective evaluation result and subjective perception is effectively improved.

Description

technical field [0001] The invention relates to an image quality evaluation method, in particular to an objective evaluation method of tone mapping image quality based on global features. Background technique [0002] With the rapid development of display technology, high dynamic range images (HDR) have attracted more and more attention. High dynamic range images have rich layers and can achieve light and shadow effects that are far closer to reality than ordinary images. However, conventional display devices can only support low dynamic range display output. In order to solve the contradiction between the dynamic range mismatch between the real scene and the traditional display device, many tone mapping (Tone Mapping) algorithms for high dynamic range images are currently proposed. The goal of the tone mapping algorithm for high dynamic range images is to compress the brightness of high dynamic range images to the acceptable range of traditional display devices, while ret...

AI Technical Summary

Problems solved by technology

[0003] For tone mapping image quality evaluation, if the existing image quality evaluation methods are directly applied to the tone mapping image, the objective evaluation value cannot be accurately predicted because the tone mapping image only has a high dynamic range image as a reference.

Images