Convolutional-neural-network-based reference-free three-dimensional image quality evaluation method

Abstract

The invention discloses a convolutional-neural-network-based reference-free three-dimensional image quality evaluation method. The method comprises the following steps: carrying out pretreatment on a 2D image; to be specific, inputting an image block obtained by pretreatment into a deep convolutional neural network, carrying out convolutional pooling processing to obtain an advanced feature of the image block, and then carrying out BP training by using the neural network based on a mass fraction of an original image so as to obtain a parameter of a base model; obtaining a corresponding basic model according to the parameter of the base model, carrying out centering on a three-dimensional image and carrying out pretreatment identical with that of the 2D image on the left image and the right image, and inputting the image blocks at the same position into the basic module simultaneously to obtain a corresponding advanced feature vector; and carrying out testing under a LIVE 3D database. On the basis of the reference-free evaluation algorithm, a good result better that the existing quality evaluation result can be obtained.

Description

technical field [0001] The invention relates to a no-reference stereoscopic image quality assessment method based on a convolutional neural network. Background technique [0002] In recent years, with the rapid development of multimedia technology, 3D stereoscopic images have entered people's lives in various forms. The realistic on-the-spot experience and shocking visual effects brought by 3D stereoscopic images are incomparable with 2D images. It not only brings people immersive visual enjoyment, but also arouses people's interest and awareness of external things. At the same time, the widespread use of stereoscopic images also puts forward higher requirements for algorithms for evaluating the quality of stereoscopic images. [0003] Stereoscopic image quality evaluation occupies a very important position in stereoscopic images. It can not only judge the pros and cons of processing algorithms in stereoscopic images, but also optimize and design algorithms to improve the e...

AI Technical Summary

Problems solved by technology

[0007] (3) No reference model: The no-reference algorithm that does not depend on the original reference image is more in line with the actual application scenario, but its implementation is also the most difficult

[0008] Therefore, there are many problems in the existing image evaluation, which can no longer meet the existing needs, and a new algorithm is urgently needed to realize the evaluation of stereoscopic image quality

Images