A Blind Stereo Image Quality Evaluation Method Based on Binocular Competition

Abstract

The invention discloses a binocular rivalry based totally blind stereo image quality evaluation method. The method comprises a training stage and a test stage; characteristics of binocular rivalry are fully utilized in both the training stage and the test stage, that is to say, binocular rivalry energy, binocular rivalry variance and binocular rivalry entropy of sub-blocks corresponding to coordinate positions in amplitude images of a left view point image and a right view point image are obtained, so that stereo visual perception characteristics are fully considered and the correlation between an objective evaluation result and subjective perception is effectively improved; an undistorted Gaussian distribution model and a distorted Gaussian distribution model are constructed in an unsupervised learning mode, so that a complicated machine learning and training process is avoided and the computing complexity is lowered; and each training distortion stereo image and a subjective evaluation value of the image do not need to be predicted in the training stage, so that the method is more suitable for actual application scenarios.

Description

technical field [0001] The invention relates to a method for evaluating the quality of a stereoscopic image, in particular to a method for evaluating the quality of a full-blind stereoscopic image based on binocular competition. Background technique [0002] Since entering the 21st century, with the maturity of stereoscopic image / video system processing technology and the rapid development of computer network and communication technology, people have a strong demand for stereoscopic image / video system. Compared with the traditional single-viewpoint image / video system, the stereoscopic image / video system is more and more popular because it can provide depth information to enhance the visual reality and give users an immersive new visual experience. It is considered to be the main development direction of the next-generation media, and has aroused widespread concern in the academic and industrial circles. However, in order to obtain better stereoscopic presence and visual exp...

AI Technical Summary

Problems solved by technology

[0003] At present, researchers have proposed many no-reference evaluation methods for single-viewpoint image quality. However, due to the lack of systematic theory to study the characteristics of stereo vision perception, there is no effective no-reference stereo image quality evaluation method.

The existing no-reference stereoscopic image quality evaluation methods mainly use machine learning to predict the stereoscopic image quality, which not only has high computational complexity, but also requires a test database (including a large number of distorted stereoscopic images of different distortion types and corresponding subjective evaluation values), This makes the no-reference stereoscopic image quality evaluation method not suitable for practical applications and has certain limitations

Images