Machine learning-based stereoscopic image quality objective assessment method

Abstract

The invention provides a machine learning-based stereoscopic image quality objective assessment method, which comprises the following steps of: extracting parameters based on which the image quality objective assessment of stereoscopic images is performed; performing machine learning by using the parameters extracted from standard image sequences in an image library; performing fitting between the stereoscopic image qualitative assessment results and the extracted parameters by using the learning results; and applying the fitting results to an image to be assessed and comparing the results with subjective assessment marks. The machine learning-based stereoscopic image quality objective assessment method provided by the invention has the beneficial effect that based on visual characteristics of human eyes, improvements on the form and weighted values of the formula for solving PSNR (Peak Signal to Noise Ratio) and SSIM (Structural Similarity ) are achieved, and a computation method of visual comfort and an assessment method using fused image parameters are set forth, and as joint fitting is performed by using a plurality of characteristic parameters, the fitting effect is improved.

Description

technical field [0001] The invention relates to the technical field of computer multimedia, in particular to an objective evaluation method of stereoscopic image quality based on machine learning. Background technique [0002] During the process of image data collection, compression processing, transmission, reconstruction, and image communication, the processed image will be different from the original image. How to evaluate these differences and the impact of these differences on image effects is the key to image quality evaluation. main content. When evaluating image quality, people are the final recipients and judges of stereoscopic images. Therefore, the stereoscopic information processing mechanism of HVS (Human Visual System) must be considered, and its perception process of stereoscopic information must be simulated. quality assessment model. [0003] At present, image quality evaluation methods can be divided into subjective evaluation methods and objective evalua...

AI Technical Summary

Problems solved by technology

Since HVS is a complex nonlinear system, it is difficult to establish an accurate evaluation method based on it; in addition, for stereoscopic images, which are different from planar images, there is a high degree of correlation between adjacent viewpoints in stereoscopic images. The image quality of adjacent viewpoints is high, but the parallax between viewpoints is small, and the observer will feel that the stereoscopic effect of the image is reduced

At present, the existing objective evaluation criteria for stereoscopic images based on human visual characteristics are all aimed at some specific distortion situations, and the applicability is not high.

Images