Author(s): |
Xiangying Mao, Faculty of Information Science and Engineering, Ningbo University, Ningbo, China Mei Yu, National Key Lab of Software New Technology, Nanjing University, Nanjing, China Junming Zhou, Faculty of Information Science and Engineering, Ningbo University, Ningbo, China Zongju Peng, Faculty of Information Science and Engineering, Ningbo University, Ningbo, China Feng Shao, Faculty of Information Science and Engineering, Ningbo University, Ningbo, China Ahong Wang, Faculty of Information Science and Engineering, Ningbo University, Ningbo, China Gangyi Jiang, National Key Lab of Software New Technology, Nanjing University, Nanjing, China |
Abstract: |
Stereoscopic image processing in wavelet domain is very critical to find a stereoscopic image qual- ity assessment model. The image quality assessment method about structure similarity proposed has excellent performance of simplicity and well correlation with subjective assessment value. In this paper, based on the research of image structural features in wavelet domain, using the visual perception characteristics, especially the band-pass characteristic of contrast sensitivity function, and combining the amplitude characteristics in wavelet domain, we propose a stereoscopic image quality assessment method which has high accordance with subjective quality score. The proposed assessment model is divided into two parts. One is the quality syn- thetic assessment of left and right perspective, and the other is stereoscopic perception assessment. The ex- perimental results show the correlation coefficient and monotonicity are above 0.90 in quality synthetic as- sessment and perception assessment for four distorted types, JPEG, JPEG2000, Gaussian blur and Gaussian white noise. It indicates the proposed stereoscopic objective model is consistent with human visual properties and is of availability.
|