No-reference stereoscopic image quality assessment method based on primary visual perception mechanism

Abstract

The invention proposes a no-reference stereoscopic image quality evaluation method based on a primary visual perception mechanism, which belongs to the technical field of image analysis. This method first converts the input stereo image pair into grayscale information, and uses the grayscale information to simulate the monocular perception response based on channel separation and the binocular perception response optimized by feedback aggregation under the primary visual perception mechanism. Secondly, the obtained monocular and binocular perceptual responses were used to simulate the classical and non-classical receptive field responses in the primary visual cortex, and the probabilistic and statistical feature vectors of the structural operator LBP were extracted from the generated two types of receptive field response images. Then, use the support vector machine to train the features to get the prediction model, and apply the prediction model and the feature vector corresponding to the test set for quality prediction and evaluation. This method has the characteristics of high subjective consistency, high database independence, and high stability. It shows very competitive effects when dealing with various complex distortion types, and has strong application value.

Description

technical field [0001] The invention relates to a method for evaluating the quality of a stereoscopic image, in particular to a method for evaluating the quality of a stereoscopic image without reference based on a primary visual perception mechanism, and belongs to the field of image analysis. Background technique [0002] In recent years, with the development of science and technology, the cost of stereoscopic image generation and dissemination has become lower and lower, which makes stereoscopic images, as an excellent medium of information dissemination, more and more popular in our daily life. Universal and increasingly indispensable. However, stereoscopic images will inevitably introduce distortion in each stage of scene acquisition, encoding, network transmission, decoding, post-processing, compression storage and projection. Blurring distortion; compression distortion caused by image compression storage, etc. The introduction of distortion will greatly reduce peopl...

AI Technical Summary

Problems solved by technology

[0018] The purpose of the present invention is to solve the problems that the simulation method of human visual perception mechanism is not perfect enough in the quality evaluation of stereoscopic images without reference, the utilization of visual perception information in images is not sufficient, the subjective consistency is poor, the independence of the database is poor, and the algorithm stability is poor. A No-Reference Stereo Image Quality Evaluation Method Based on Primary Visual Perception Mechanism

Images