Human-visual-system (HVS)-based structural similarity (SSIM) and characteristic matching three-dimensional image quality evaluation method

Abstract

The invention relates to image quality evaluation. To show the fidelity and third dimension of a generated three-dimensional image, the degree of damages of a compression algorithm to the three-dimensional image, the degree of interference of noises introduced by a transmission process on the quality of the three-dimensional image, the display naturalness of the three-dimensional image, and the like, the technical scheme adopted by the invention is that: a human-visual-system (HVS)-based structural similarity (SSIM) and characteristic matching three-dimensional image quality evaluation methodcomprises the following steps of: (1) comparing the luminance, contrast and structural similarity of left and right views of an original image with those of the left and right views of a test image by using a structure distortion method; (2) extracting luminance and contrast indexes; (3) simulating a human eye band-pass property principle according to wavelet decomposition to obtain a human visual signal to noise ratio evaluation index; (4) reflecting the third dimension of the three-dimensional image by using the ratio of number of left and right view matching points of the test image to thenumber of the left and right view matching points of the original image; and (5) rationally weighting all the indexes to obtain an overall evaluation index. The method is mainly applied to the image quality evaluation.

Description

Technical field [0001] The invention relates to image quality evaluation and 3D image processing, in particular to an HVS-based SSIM and feature matching stereo image quality evaluation method. Background technique [0002] Pieter proposed a subjective quality evaluation method for 3DTV, pointing out that the evaluation should be made in terms of image quality, image depth, three-dimensional comfort, whether three-dimensional feeling exists, whether three-dimensional feeling is natural, whether three-dimensional feeling conforms to visual experience, etc.; Patrizio proposed a stereo image Subjective quality evaluation method. Subjects wear stereo glasses, use France Telecom R&D's SEOVQ (Subjective Evaluation and Optimization of Video Quality) human-computer interface for subjective evaluation, and use the average difference score (DMOS) to represent the evaluation result. Subjective evaluation methods are time-consuming, labor-intensive, and have certain instability factors; Patr...

AI Technical Summary

Problems solved by technology

The subjective evaluation method is time-consuming, labor-intensive, and has certain unstable factors; Patrizio et al. proposed to combine multiple two-dimensional image evaluation indicators SSIM (structural similarity index), UQI (Universal Quality Index traditional quality index), RRIQA ( Reduce the quality evaluation of the reference image) for the evaluation of the stereoscopic image, on the basis of evaluating the quality of the left and right images respectively, the evaluation result is given by the method of weighted summation

This method has reference value, but does not consider the corresponding stereoscopic vision mechanism; Alexandre proposed an objective quality evaluation method for stereoscopic images based on parallax, which calculated the plane image quality distortion and parallax distortion respectively, and then weighted the two, the parallax The distortion is calculated by the correlation coefficient between the original image pair and the test image pair depth map. The plane image quality distortion uses the plane image evaluation method to calculate the distortion index between the left image and the right image respectively, and then calculates the average value. However, in Alexandre’s In the method, the correlation coefficient between the original image pair and the test image pair depth map is used to characterize the stereoscopic information characteristics, which can only give an overall understanding from the perspective of the whole image, and cannot clearly reflect the impact of local parallax distortion on image quality evaluation. ;Ningbo University proposed a method for evaluating the quality of stereoscopic right images based on similarity. This method uses parallax in the quality evaluation of right images. In essence, it only evaluates the quality of single-viewpoint images, not the real stereoscopic image quality evaluation.

Images