Non-reference 3D stereo image quality evaluation method based on asymmetric distortion

Abstract

The invention relates to a non-reference 3D stereoscopic image quality evaluation method based on asymmetric distortion, comprising the following steps: decomposing a stereoscopic image pair by using a Gabor filter group to obtain Gabor pyramids of left view and right view respectively. The fused cyclopean image is obtained by adding pixel-by-pixel points of the left-view and right-view Gabor pyramids. The difference map is generated by subtracting the absolute difference values ​​of the left and right view Gabor pyramids. The generalized Gaussian model GGD is used to fit the cyclopean diagram and the difference diagram, and the statistical characteristic parameters after fitting are extracted, including the mean value, variance and shape parameters of the GGD distribution. Calculate the structural similarity SSIM score between the left view and the right view decomposed by the Gabor filter, as the asymmetric information of the stereo image, combined with the statistical feature parameters extracted in the previous step, use support vector regression SVR to predict the image quality score .

Description

Technical field: [0001] The invention relates to the field of objective quality evaluation for 3D stereo images. Background technique: [0002] With the rapid development of science and technology and the upgrading of science and technology, products in the digital age have gradually entered people's lives. Under the tide of the digital age, the field of 3D stereoscopic images is developing and applied rapidly. Compared with two-dimensional flat images, stereoscopic images can enable people to experience the sense of three-dimensionality and presence, and gradually become the mainstream of multimedia research. direction. However, during the process of encoding, compressing, and transmitting stereoscopic images, different forms of distortion often exist, and all of these distortions will cause a decrease in visual quality. Therefore, how to accurately and effectively evaluate image quality has become a research hotspot in the field of image processing. [0003] The 3D ster...

AI Technical Summary

Problems solved by technology

[0005] The technical problem solved by the present invention: Aiming at the distortion of asymmetric 3D images, a new no-reference 3D stereoscopic image quality evaluation method is proposed, a Gabor filter is designed to perform pyramid decomposition on the left and right views of the stereoscopic image, and according to the principle of binocular competition and binocular fusion, the The left and right images are fused to obtain the Cyclops image and the difference image, and then the two obtained images are normalized, then the generalized Gaussian model is fitted, and the statistical features are extracted respectively. Finally, the support vector regression SVR is used to predict the quality score. The experimental results show that The proposed SIQA algorithm achieves very good results

Images