A No-reference Stereo Image Quality Evaluation Method

Abstract

The invention discloses a method for evaluating the quality of a stereoscopic image without reference. In the training stage, the training image set composed of the original left viewpoint image and the distorted left viewpoint image, the original right viewpoint image and the distorted right viewpoint image are obtained through joint dictionary training. The image feature dictionary table and the image quality dictionary table of the training image set composed of the training image set, the original Cyclops image and the distorted Cyclops image; no need to calculate the image feature dictionary table and the image quality dictionary table in the test phase, avoiding the complexity machine learning training process, and there is no need to predict the subjective evaluation value of each test distorted stereo image; in the test phase, according to the image feature dictionary table and image quality dictionary table constructed in the training phase, and then considering the left viewpoint image of the test distorted stereo image , the weight coefficients of the right view image and the cyclops image to obtain the predicted value of the objective evaluation of image quality, which can improve the correlation between the objective evaluation result and the subjective perception.

Description

technical field [0001] The invention relates to an image quality evaluation method, in particular to a reference-free stereoscopic image quality evaluation method. Background technique [0002] With the rapid development of image coding and display technologies, the research on image quality evaluation has become a very important link. The objective of the research on the objective evaluation method of image quality is to be consistent with the subjective evaluation results as much as possible, so as to get rid of the time-consuming and boring subjective evaluation method of image quality, which can automatically evaluate the image quality by computer. According to the degree of reference and dependence on the original image, objective image quality evaluation methods can be divided into three categories: full reference (FR) image quality evaluation methods, partial reference (Reduced Reference, RR) image quality evaluation methods and no reference ( No Reference, NR) image...

AI Technical Summary

Problems solved by technology

At present, the existing no-reference image quality evaluation method uses machine learning to predict the evaluation model, but its computational complexity is high, and the training model needs to predict the subjective evaluation value of each evaluation image, which is not suitable for practical applications , there are certain limitations

Images