Measurement method of 3D image quality based on deep learning

Abstract

The invention belongs to the field of image processing, and aims to propose a new stereoscopic image quality evaluation method, realize more accurate and effective evaluation of stereoscopic image quality, and promote the development of stereoscopic imaging technology to a certain extent. For this reason, the technical solution adopted by the present invention is, based on the measurement method of 3D image quality of deep learning, firstly carry out block processing on the stereoscopic image data set, obtain many small image blocks through the block processing, and then carry out each image block Normalization processing; at the same time, principal component analysis PCA dimensionality reduction processing is performed on the stereoscopic image dataset to obtain images with lower dimensions; then the image block dataset obtained by cutting and the low-dimensional dataset obtained after PCA dimensionality reduction are sent to the In the built convolutional neural network; then use the convolutional neural network to extract features layer by layer; finally get the overall quality of the stereo image through the softmax classifier. The invention is mainly applied to image processing.

Description

technical field [0001] The invention belongs to the field of image processing, and relates to the improvement and optimization of a stereoscopic image quality evaluation method, in particular to the application of deep learning convolutional neural network and principal component analysis in the objective evaluation of stereoscopic image quality. Background technique [0002] With the rapid development of mobile devices and communications, people are exposed to more and more picture content in their lives, especially the recently emerging 3D display technology and related applications have greatly improved the visual experience of the human eye, such as 3D movies, VR glasses, etc., have brought more entertainment and unique experience, which has attracted more researchers not only in the industry, but also in academia. How to effectively evaluate the quality of stereoscopic images in real time has become a hot topic in the field of stereoscopic image research. one of the key...

Images