A method and system for subjective evaluation of video image quality

Abstract

The invention belongs to the field of computer digital video image processing, and discloses a method and system for subjective evaluation of video image quality. The evaluation method includes: determining evaluators, scoring standards and viewing conditions, selecting test sequences that meet the requirements; for the selected test sequence group, using the evaluation system to test; saving test results, and performing data statistics and analysis. The evaluation system realizes the recording of evaluator information, data statistics and analysis of evaluation results, and controls the entire evaluation process, including: a user management module, a material management module, a sequence playback module, and a data management module. The invention effectively combines three evaluation modes of multi-stimulus evaluation mode, double-stimulus-like continuous quality evaluation mode and dual-screen display to subjectively evaluate video image quality, and expands the scope of application of the subjective evaluation method. When evaluating qualitatively indistinguishable sequences, the present invention improves the stability and reliability of the evaluation results.

Description

technical field [0001] The invention belongs to the field of computer digital video image processing, and relates to a method and system for subjective evaluation of video image quality. Background technique [0002] With the development of video services, people have higher and higher requirements for video image quality. Because video image data will inevitably affect the image quality in the process of acquisition, compression, transmission, processing, storage, and recovery, the accurate evaluation of video image quality has become one of the key technologies in the field of video image research. Evaluation methods of video image quality can be divided into two categories: objective evaluation and subjective evaluation. Objective evaluation methods are easy to operate and highly repeatable, but there may be problems that the evaluation results are not completely consistent with people's subjective feelings. The subjective evaluation method is more direct and reliable, ...

AI Technical Summary

Problems solved by technology

[0005] Most of the subjective evaluations of video image quality in the above literatures use a single subjective evaluation method and corresponding software systems, and hardly involve the combination of two or more subjective evaluation methods.

However, on the basis of referring to a large number of literature and experimental experience, it is found that several commonly used subjective evaluation methods have certain limitations in the scope of application.

For example, when the image quality damage is not significantly different, since the multi-stimulus method allows the evaluators to repeatedly watch the source reference sequence and the distortion test sequence, more accurate evaluation results can be obtained compared with the single-stimulus method and the double-stimulus method; When the image quality impairment is very different or it is impossible to test the distorted image and the source reference image together, the single-stimulus method can quickly and efficiently obtain the evaluation results; in addition, when evaluating the relative quality of two distortion test sequences, the current evaluation Algorithms are hard to implement

Therefore, there is a certain degree of insufficiency in the existing literature on the subjective evaluation of video image quality, so that the evaluation results may lack reliability and stability

Images