New objective assessment model of stereo images based on human visual system

被引：0

作者：

Zhu, Jiangying ^{[1
]}

Yu, Mei ^{[1
]}

Jiang, Gangyi ^{[1
]}

Peng, Zongju ^{[1
]}

Shao, Feng ^{[1
]}

Zhou, Junming ^{[1
]}

机构：

[1] Faculty of Information Science and Engineering, Ningbo University, Ningbo

来源：

Journal of Convergence Information Technology | 2012年 / 7卷 / 15期

关键词：

Human visual system; Quality assessment; Stereo images; Stereo perception;

D O I：

10.4156/jcit.vol7.issue15.23

中图分类号：

学科分类号：

摘要：

Stereoscopic image quality assessment is becoming a key issue in stereo image processing technologies. Since human beings are the final receivers of stereo images, an objective stereo image quality assessment model based on the characteristics of human visual system is proposed in this paper. The model is divided into two parts: quality assessment of two view images and stereo perception assessment. The former is firstly established on the basis of the characteristics of luminance nonlinearity, and then the structural similarities of texture, flat, and edge regions are calculated. The later applies the wavelet domain of the absolute disparity image to simulate multi-channel effects, uses the contrast sensitivity function to weight the different spatial frequencies, and assesses the stereo perception of stereo images through human visual signal noise ratio. Finally, the two parts are combined to represent the quality index of stereo images. Experimental results show that the predictive results of the proposed model are consistent with subjective evaluations. Both the correlation coefficient and monotonous are above 0.92, and the mean square errors are less than 6.5 for five different distortion types. These three indices indicate that the proposed model can predict human visual perception well.

引用

页码：194 / 202

页数：8

共 17 条

[1]

Abdelouahad A.A., Hassouni M.E., Cherifi H., Aboutajdine D., A reduced reference image quality measure using Bessel K forms model for Tetrolet coefficients, Journal of Convergence Information Technology, 6, 11, pp. 216-224, (2011)

[2]

Utriainen T., Hayrynen J., Jumisko-Pyykko S., Boev A., Gotchev A., Hannuksela M.M., Mobile 3D quality of experience evaluation: A hybrid data collection and analysis approach, Proc. of SPIE-IS&T Electronic Imaging, SPIE, 7881, (2011)

[3]

Maalouf A., Larabi M.-C., CYCLOP: A stereo color image quality assessment metric, Int. Conf. on Audio Speech Signal Processing, pp. 1161-1164, (2011)

[4]

Goldmann L., Ebrahimi T., Towards reliable and reproducible 3D video quality assessment, Proc. of SPIE, 8043, (2011)

[5]

Ding Y., Zhang Y., Zhang D.Z.D., Wang X., Weighted multi-scale structural similarity for image quality assessment with saliency-based pooling strategy, International Journal of Digital Content Technology and its Applications, 6, 5, pp. 67-78, (2012)

[6]

Jiang G., Huang D., Wang X., Yu M., Qverview on image quality assessment methods, Journal of Electronics & Information Technology, 32, 1, pp. 219-226, (2010)

[7]

Benoit A., Callet P.L., Campisi P., Cousseau R., Using disparity for quality assessment of stereoscopic images, Int. Conf. on Image Processing, pp. 389-392, (2008)

[8]

Shao H., Cao X., Er G., Objective quality assessment of depth image based rendering in 3DTV system, 3DTV Conference, Potsdam, Germany, pp. 1-4, (2009)

[9]

Ma L., Li S., Zhang F., Reduced-reference image quality assessment using reorganized DCT-based image representation, IEEE Transactions on Multimedia, 13, 4, pp. 824-829, (2011)

[10]

Li S., Xiang W., Cheng F., Zhao R., Hou C., HVS-based quality assessment metrics for 3-D images, WRI Global Congress on Intelligent Systems, pp. 86-89, (2010)

← 1 2 →