Multimedia Tools and Applications

, Volume 77, Issue 10, pp 12139–12156 | Cite as

Saliency-based adaptive compressive sampling of images using measurement contrast

  • Ran Li
  • Wei He
  • Zhenghui Liu
  • Yanling Li
  • Zhangjie Fu


Compressive Sampling (CS) achieves the sub-Nyquist image acquisition, which bringing about a rapid development of compressive imaging devices. In CS framework, the adaptive sampling scheme is an efficient approach to improving the rate-distortion performance of imaging system. However, the sampling allocation depends on the original sample image, which increases the cost and complexity of imaging system, thereby making CS lose its superiority. In this paper, we propose a saliency-based adaptive CS scheme that allocates more sampling resources to salient regions but fewer to non-salient regions. Its key idea is to extract the saliency information by using the contrast between CS measurements, thus avoiding the original sample image in the imaging system. The scheme is realized in practice without any changes of the architecture of compressive imaging device. To match our adaptive sampling scheme, we also propose a weighted global recovery model based on saliency information. This model can effectively suppress the blocking artifacts while improving the visual qualities of salient regions. Experimental results on natural images show that the proposed adaptive CS scheme improves the visual quality of reconstructed image, and has better rate-distortion performance than the existing adaptive CS schemes.


Compressive sampling Saliency detection Measurement contrast Adaptive sampling Weighted global recovery 



This work was supported in part by the National Natural Science Foundation of China, under Grants nos. 61501393, 61601396, 61572417 and 61502409, in part by Youth Sustentation Fund of Xinyang Normal University, under Grant no. 2015-QN-043, in part by the Key Scientific Research Project of Colleges and Universities in Henan Province of China, under Grant no. 16A520069.


  1. 1.
    Candè EJ, Wakin MB (2008) An introduction to compressive sampling. IEEE Signal Process mag 25(2):21–30CrossRefGoogle Scholar
  2. 2.
    Chen Y, Hao C, Wu W, Wu E (2016) Robust dense reconstruction by range merging based on confidence estimation. SCIENCE CHINA Inf Sci 59(9):1–11Google Scholar
  3. 3.
    Duarte MF, Davenport MA, Takbar D, Laska JN, Sun T, Kelly KF, Baraniuk RG (2008) Single-pixel imaging via compressive sampling. IEEE Signal Process Mag 25(2):83–91CrossRefGoogle Scholar
  4. 4.
    Figueiredo MAT, Nowak RD, Wright SJ (2007) Gradient projection for sparse reconstruction: application to compressed sensing and other inverse problems. IEEE J Sel Top Sign Proces 1(4):586–597CrossRefGoogle Scholar
  5. 5.
    Gan L (2007) Block compressed sensing of natural images. In: Proceedings of 15th IEEE International Conference on Digital Signal Processing, pp 403–406Google Scholar
  6. 6.
    Itti L, Koch C (2001) Computational modelling of visual attention. Nat Rev Neurosci 2(3):194–203CrossRefGoogle Scholar
  7. 7.
    Itti L, Koch C, Niebur E (1998) A model of saliency-based visual attention for rapid scene analysis. IEEE Trans Pattern Anal Mach Intell 20(11):1254–1259CrossRefGoogle Scholar
  8. 8.
    Li R, Gan Z, Cui Z, Wu M, Zhu X (2013) Distributed adaptive compressed video sensing using smoothed projected Landweber reconstruction. China Commun 10(11):58–69CrossRefGoogle Scholar
  9. 9.
    Muhammad K, Ahmad J, Sajjad M, Baik SW (2016) Visual saliency models for summarization of diagnostic hysteroscopy videos in healthcare systems. SpringerPlus 5(1):1495CrossRefGoogle Scholar
  10. 10.
    Muhammad K, Sajjad M, Mi YL, Baik SW (2017) Efficient visual attention driven framework for key frames extraction from hysteroscopy videos. Biomed Signal Proces Control 33:161–168CrossRefGoogle Scholar
  11. 11.
    Pan Z, Zhang Y, Kwong S (2015) Efficient motion and disparity estimation optimization for low complexity multiview video coding. IEEE Trans Broadcast 61(2):166–176CrossRefGoogle Scholar
  12. 12.
    Pan Z, Jin P, Lei J, Zhang Y, Sun X, Kwong S (2016) Fast reference frame selection based on content similarity for low complexity HEVC encoder. J Vis Commun Image Represent 40(Part B):516–524CrossRefGoogle Scholar
  13. 13.
    Pan Z, Lei J, Zhang Y, Sun X, Kwong S (2016) Fast motion estimation based on content property for low-complexity H.265/HEVC encoder. IEEE Trans Broadcast 62(3):675–684CrossRefGoogle Scholar
  14. 14.
    Stankovi V, Stankovi L, Cheng S (2009) Compressive image sampling with side information. In: Proceedings of IEEE International Conference on Image Processing, pp 3037–3040Google Scholar
  15. 15.
    Tan J, Ma Y, Rueda H, Baron D, Arce GR (2016) Compressive hyperspectral imaging via approximate message passing. IEEE J Sel Top Sign Proces 10(2):389–401CrossRefGoogle Scholar
  16. 16.
    Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600–612CrossRefGoogle Scholar
  17. 17.
    Wu X, Dong W, Zhang X, Shi G (2012) Model-assisted adaptive recovery of compressed sensing with image applications. IEEE Trans Image Process 21(2):451–458MathSciNetCrossRefMATHGoogle Scholar
  18. 18.
    Yu Y, Wang B, Zhang L (2010) Saliency-based compressive sampling for image signals. IEEE Signal Process Lett 17(11):973–976CrossRefGoogle Scholar
  19. 19.
    Zhang J, Zhao D, Xiong R et al (2014) Group-based sparse representation for image restoration. IEEE Trans Image Process 23(8):3336–3351MathSciNetCrossRefMATHGoogle Scholar
  20. 20.
    Zhang J, Xiang Q, Yin Y, Chen C, Luo X (2017) Adaptive compressed sensing for wireless image sensor networks. Multimedia Tools Appl 76(3):4227–4242CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  1. 1.School of Computer and Information TechnologyXinyang Normal UniversityXinyangChina
  2. 2.School of Computer and SoftwareNanjing University of Information Science & TechnologyNanjingChina
  3. 3.Shenzhen Key Laboratory of Media SecurityShenzhen UniversityShenzhenChina

Personalised recommendations