Journal of the Institute of Electronics and Information Engineers (전자공학회논문지)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지
DOI QR코드

DOI QR Code

Compressive Sensing Recovery of Natural Images Using Smooth Residual Error Regularization

평활 잔차 오류 정규화를 통한 자연 영상의 압축센싱 복원

  • Trinh, Chien Van (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Dinh, Khanh Quoc (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Nguyen, Viet Anh (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Park, Younghyeon (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Jeon, Byeungwoo (College of Information and Communication Engineering, Sungkyunkwan University)
  • ;
  • ;
  • ;
  • 박영현 (성균관대학교 정보통신대학) ;
  • 전병우 (성균관대학교 정보통신대학)
  • Received : 2014.02.19
  • Accepted : 2014.06.05
  • Published : 2014.06.25
Download PDF
⟨ Previous Next ⟩

Abstract

Compressive Sensing (CS) is a new signal acquisition paradigm which enables sampling under Nyquist rate for a special kind of signal called sparse signal. There are plenty of CS recovery methods but their performance are still challenging, especially at a low sub-rate. For CS recovery of natural images, regularizations exploiting some prior information can be used in order to enhance CS performance. In this context, this paper addresses improving quality of reconstructed natural images based on Dantzig selector and smooth filters (i.e., Gaussian filter and nonlocal means filter) to generate a new regularization called smooth residual error regularization. Moreover, total variation has been proved for its success in preserving edge objects and boundary of reconstructed images. Therefore, effectiveness of the proposed regularization is verified by experimenting it using augmented Lagrangian total variation minimization. This framework is considered as a new CS recovery seeking smoothness in residual images. Experimental results demonstrate significant improvement of the proposed framework over some other CS recoveries both in subjective and objective qualities. In the best case, our algorithm gains up to 9.14 dB compared with the CS recovery using Bayesian framework.

압축센싱은 성긴 (sparse) 신호에 대해 Nyquist rate 미만의 샘플링으로도 신호 획득이 가능하다는 것을 수학적으로 증명한 새로운 개념이다. 그동안 영상분야 압축센싱을 위한 수많은 복원 알고리즘들이 제안되어 왔으나, 낮은 측정률 하에서는 복원 화질 측면에서 아직 개선할 점이 많다. 일례로, 자연 영상의 압축센싱 복원 화질 향상을 위해, 영상과 관련한 사전 정보들로부터 정규화 식을 도출하여 복원에 적용해 볼 수 있을 것이다. 따라서, 본 논문에서는 Dantzig selector 및 평활 필터(가우시안 필터 및 nonlocal 평균 필터)기반의 평활 잔차 오류 정규화 방법을 제안한다. 또한, 복원 영상의 객체 및 배경에서 발생하는 edge 정보를 우수하게 보전하는 것으로 알려진 Total variation 기반 최소화 알고리즘에 적용하여 복원 영상의 화질을 향상시키는 방법을 제안한다. 제안하는 구조는 잔차신호의 평활화를 활용한다는 측면에서 새로운 압축센싱 복원 방식이라고 할 수 있다. 실험 결과, 제안방법은 기존 방법들에 비해 객관적 및 주관적 화질 측면에서 더 높은 성능 향상을 보여주었으며, 특히 기존 Bayesian 압축센싱 복원 방식과 비교 시 최대 9.14 dB 성능이 향상되었다.

Keywords

References

  1. D. L. Donoho, "Compressive Sensing," IEEE Trans. on Inform. Theory., vol. 52, no. 4, pp. 1289-1306, Apr. 2006. https://doi.org/10.1109/TIT.2006.871582
  2. S. Park, H. N. Lee, and S. Park, "Introduction to Compressed Sensing," The Magazine of the IEEK, vol. 38, no. 1, pp. 19-30, 2011.
  3. Y. M. Cho, "Compressive Sensing-Mathematical Principles and Practical Implications-," The Magazine of the IEEK, vol. 38, no. 1, pp. 31-43, 2011.
  4. Y. C. Eldar and G. Kutyniok, Compressed Sensing: Theory and App., Cambridge Univ. Press, 2012.
  5. D. Lee, K. Kim, C.-B. Ahn, and H. Park, "Adaptive Measurement for Performance Improvement of Compressed Sensing," Journal of The Institute of Electronics of Korea, vol. 49, no. 9, pp. 85-91, 2012. https://doi.org/10.5573/ieek.2012.49.9.085
  6. L. He, H. Chen, and L. Carin, "Tree-structured Compressive Sensing with variational Baysian analysis," IEEE Signal Process. Letters, vol. 17, no. 3, pp. 233-236, 2010. https://doi.org/10.1109/LSP.2009.2037532
  7. S. Mun and J. E. Flower, "Block Compressed Sensing of Images using Directional Transform," in Proc. of IEEE Intern. Conf. on Image Process., pp. 3021-3024, Nov. 2009.
  8. C. V. Trinh, K. Q. Dinh, and B. Jeon, "Edge-preserving Block Compressive Sensing with Projected Landweber," in Proc. of IEEE Intern. Conf. on Systems, Signals and Image Process. , pp. 71-73, Jul. 2013.
  9. M. Afonso and et al. "Fast Image Recovery Using Variable Splitting and Constrained Optimization," IEEE Trans. on Image Process., vol. 19, no. 9, pp. 2345-2356, Sep. 2010. https://doi.org/10.1109/TIP.2010.2047910
  10. J. Xu and et al., "Improved Total Variation Minimization Method for Compressive Sensing by Intra-Prediction," Signal Process., vol. 20, no. 11, pp. 2378-2386, Aug. 2011.
  11. C. Li, W. Yin, H. Jiang, and Y. Zhang, "An Efficient Augmented Lagrangian Method with Applications to Total Variation Minimization," Computational and Applied Mathematics, vol. 56, issue. 3, pp. 507-530, Dec. 2013.
  12. C. V. Trinh, K. Q. Dinh, V. A, Nguyen, B. Jeon, and D. Sim, "Compressive Sensing Recovery with Improved Hybrid Filter," in Proc. of. IEEE Inter. Congress on Image and Sig. Process., Dec. 2013.
  13. E. Candes and T. Tao, "The Dantzig selector: Statistical Estimation when p is much larger than n," Annals of Statistics Jour., vol. 6, no. 6, pp. 2313-2351, 2007.
  14. J. M. Geusebroek, et al, "Fast Anisotropic Gauss filtering," IEEE Trans. on Image Process., vol. 12, no. 8, pp. 938-943, Aug. 2003. https://doi.org/10.1109/TIP.2003.812429
  15. L. Zhang, L. Zhang, X. Mou, and D. Zhang, "FSIM: A Feature SIMilarity index for Image Quality Assessent," IEEE Trans. Image Process., vol. 20, issue. 8, pp. 2378-2386, Aug. 2011. https://doi.org/10.1109/TIP.2011.2109730
  16. K. Q. Dinh, H. J. Shim, and B. Jeon, "Deblocking Filter for Artifact Reduction in Distributed Compressive Video Sensing," in Proc. of IEEE Visual Com. and Image Process., pp. 1-5, Nov. 2012.
  17. M. R. Hestenes, "Multiplier and Gradient Methods," Jour. of Opt. Theory and Appl., vol. 4, no. 5, pp. 303-320, 1969. https://doi.org/10.1007/BF00927673
  18. A. Buades and et. al, "A Nonlocal Algorithm for Image Denoising," IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 60-65, Jun. 2005.
  19. R. C. Gonzales and R. E. Woods, Digital Image Processing, 3rd ed., Prentice Hall, New Jersey, Aug. 2007.
  20. V. Estellers, J. Thiran, and X. Bresson, "Enhanced Compressed Sensing Recovery with Level Set Normals," IEEE Trans. on Image Process., vol. 22, no. 7, pp. 2611-2626, Jul. 2013. https://doi.org/10.1109/TIP.2013.2253484