Visual saliency induced local image quality metrics

被引：0
|
作者：
Gao M. [1 ]
Dang H. [1 ]
Wei L. [2 ]
Zhang X. [1 ]
机构：
[1] College of Electrical and Information Engineering, Shaanxi University of Science & Technology, Xi'an
[2] School of Mathematics and Statistics, Ningxia University, Yinchuan
来源：
Zhongguo Kexue Jishu Kexue/Scientia Sinica Technologica | 2019年 / 49卷 / 11期
关键词：
Edge strength; Image quality assessment; Local image quality; Visual saliency;
D O I：
10.1360/N092018-00438
中图分类号：
学科分类号：
摘要：
The full reference image quality assessment usually includes two stages: local image quality measurement and pooling. The classical assessment algorithm based on visual saliency (VS) applies VS in the pooling stage and uses VS to weighted average the similarity map. It emphasizes that the region with strong significance has a large contribution to the overall image quality, while the region with weak significance has a small contribution to the overall image quality. Different from the classical method, VS is used in the measurement stage of local image quality in this paper, and VS is used to adjust the calculation of local image quality adaptively, emphasizing that the local quality degradation perceived by human visual system is jointly determined by objective degradation degree and significance. The main contributions of this paper include: (1) propose a local image quality measurement framework guided by visual saliency; (2) within this framework, we further improve the ESSIM (edge strength similarity) algorithm previously published on IEEE Signal Processing Letters, and propose the VS-guided ESSIM algorithm (VS-ESSIM). The experimental results on TID2013, TID2008 and CSIQ show that the proposed algorithm can improve the prediction accuracy of image quality and achieve better consistency with the subjective evaluation results (https://github. com/zhangprofessor/Visual-saliency-induced-local-imagequality- metrics). © 2019, Science Press. All right reserved.
引用
收藏
页码：1350 / 1360
页数：10
相关论文
共 24 条
  • [1] Mohammadi P., Ebrahimi-Moghadam A., Shirani S., Subjective and objective quality assessment of image: A survey, Maj J Electr Eng, 9, pp. 419-423, (2015)
  • [2] Pang K., Shi Z., Li Z., An image quality assessment index based on visual saliency and gradient amplitude for telemedicine application, Proceedings of 2017 International Conference on Information Science and Control Engineering, pp. 172-176, (2017)
  • [3] Nan D., Bi D.Y., Ma S.P., Et al., A quality assessment method with classified-learning for dehazed images, Acta Autom Sin, 42, pp. 270-278, (2016)
  • [4] Chu J., Chen Q., Yang X.C., Review on full reference image quality assessment algorithms, Appl Res Comput, 31, pp. 13-22, (2014)
  • [5] Jiang G.Y., Huang D.J., Wang X., Et al., Overview on image quality assessment methods, J Electron Inf Technol, 32, pp. 219-226, (2010)
  • [6] Wang Z., Bovik A.C., Mean squared error: Love it or leave it? A new look at signal fidelity measures, IEEE Signal Process Mag, 26, pp. 98-117, (2009)
  • [7] Huynh-Thu Q., Ghanbari M., Scope of validity of PSNR in image/video quality assessment, Electron Lett, 44, pp. 800-801, (2008)
  • [8] Wang Z., Bovik A.C., Sheikh H.R., Et al., Image quality assessment: from error visibility to structural similarity, IEEE Trans Image Process, 13, pp. 600-612, (2004)
  • [9] Liu H.T., Heynderickx I., Visual attention in objective image quality assessment: Based on eye-tracking data, IEEE Trans Circuits Syst Video Technol, 21, pp. 971-982, (2011)
  • [10] Liu H.T., Engelke U., Junle Wang U., Et al., How does Image Content Affect the Added Value of Visual Attention in Objective Image Quality Assessment, IEEE Signal Process Lett, 20, pp. 355-358, (2013)
123