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

Smoothed Group-Sparsity Iterative Hard Thresholding Recovery for Compressive Sensing of Color Image

컬러 영상의 압축센싱을 위한 평활 그룹-희소성 기반 반복적 경성 임계 복원

  • Nguyen, Viet Anh (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Dinh, Khanh Quoc (College of Information and Communication Engineering, Sungkyunkwan University) ;
  • Van Trinh, Chien (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 : 2014.02.23
  • Accepted : 2014.04.02
  • Published : 2014.04.25
Download PDF
⟨ Previous Next ⟩

Abstract

Compressive sensing is a new signal acquisition paradigm that enables sparse/compressible signal to be sampled under the Nyquist-rate. To fully benefit from its much simplified acquisition process, huge efforts have been made on improving the performance of compressive sensing recovery. However, concerning color images, compressive sensing recovery lacks in addressing image characteristics like energy distribution or human visual system. In order to overcome the problem, this paper proposes a new group-sparsity hard thresholding process by preserving some RGB-grouped coefficients important in both terms of energy and perceptual sensitivity. Moreover, a smoothed group-sparsity iterative hard thresholding algorithm for compressive sensing of color images is proposed by incorporating a frame-based filter with group-sparsity hard thresholding process. In this way, our proposed method not only pursues sparsity of image in transform domain but also pursues smoothness of image in spatial domain. Experimental results show average PSNR gains up to 2.7dB over the state-of-the-art group-sparsity smoothed recovery method.

압축센싱은 성긴(Sparse) 또는 압축가능한(Compressible) 신호에 대해 Nyquist rate 미만의 샘플링으로도 신호 복원이 가능하다는 것을 수학적으로 증명한 새로운 패러다임의 신호 획득 방법이다. 단순한 신호 획득 과정을 이용하면서도, 동시에 우수한 압축센싱 복원 영상을 얻기 위한 많은 연구들이 수행되고 있다. 그러나, 에너지 분포 및 인간 시각 시스템 등 컬러 영상에 대한 기본적인 특성을 복원 과정에 활용한 기존 압축센싱 관련 연구는 많이 부족하다. 이러한 문제를 해결하기 위해, 본 논문에서는 컬러영상의 압축센싱 복원을 위한 평활 그룹-희소성 기반 반복적 경성 임계 알고리즘을 제안한다. 제안하는 방법은 그룹-희소성에 기반한 경성 임계치 적용과 프레임 기반 필터의 사용을 통해 영상의 변환 영역에 대한 희소성을 증대시키는 동시에 화소 영역의 평활 정도를 복원 과정에 활용할 수 있도록 한다. 또한, 그룹-희소화 경성 임계 과정은 자연 영상의 에너지 분포 및 인간 시각시스템 특성에 따라 중요하다고 판단되는 RGB-그룹 계수들을 보전하도록 설계하였다. 실험 결과 객관적 화질 측면에서 제안방법이 대표적인 그룹-희소화 평활 복원 기법 보다 평균 PSNR이 최대 2.7dB 높은 것을 확인하였다.

Keywords

References

  1. C. E. Shannon, "Communication in the presence of noise", Proc. Institute of Radio Engineers, vol. 37, no.1, pp. 10-21, Jan. 1949. Reprint as classic paper in: Proc. IEEE, vol. 86, No. 2, Feb. 1998.
  2. R. G. Baraniuk, "Compressive Sensing", IEEE Signal Process. Magazine, vol. 24, pp. 118-121, Jul. 2007.
  3. S. Park, H. N. Lee, S. Park, "Introduction to Compressed Sensing," The Magazine of the IEEK, vol. 38, no. 1, pp. 19-30, 2011.
  4. Y. M. Cho, "Compressive Sensing-Mathematical Principles and Practical Implications," The Magazine of the IEEK, vol. 38, no. 1, pp. 31-43, 2011.
  5. A. Majumdar and R. K. Ward, "Compressed sensing of color images," Signal Processing, vol. 90, pp. 3122-3127, Dec. 2010. https://doi.org/10.1016/j.sigpro.2010.05.016
  6. P. Nagesh and B. Li, "Compressive imaging of color images," IEEE Int. Conf. Acoust., Speech Signal Processing (ICASSP), pp. 1261-1264, Apr. 2009.
  7. W. Kang, E. Lee, S. Kim, J. Paik, "Noise removing method using compressive sensing," 2011 IEEK Summer Conference, vol. 34, no. 1, pp. 1676-1679, 2011.
  8. M. Wakin et al., "An architecture for compressive imaging," IEEE Int. Conf. Image Processing (ICIP), pp. 1273-1276, Oct. 2006.
  9. V. A. Nguyen, K. Q. Dinh, C. V. Trinh, and B. Jeon, "Visually weighted groups-sparsity recovery for compressive sensing of color images with edge-preserving filter," IEEE Int. Conf. Image Processing (ICIP), 2014 (submitted).
  10. T. Blumensath and M. E. Davies, "Iterative thresholding for sparse approximation," Jour. Fourier Analysis and Applications, vol. 14, pp. 629-654, Dec. 2008. https://doi.org/10.1007/s00041-008-9035-z
  11. H. T. Kung and S. J. Tarsa, "Partitioned compressive sensing with neighbor-weighted decoding," Int. Conf. Military Comm. (MILCOM), pp. 149-156. Nov. 2011.
  12. 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
  13. E. Candes, J. Romberg, and T. Tao, "Decoding by linear programming," IEEE Trans. Inform. Theory, vol. 51, pp. 4203-4215, Dec. 2005. https://doi.org/10.1109/TIT.2005.858979
  14. E. Candes and J. Romberg, "Sparsity and Incoherence in Compressive Sampling," Inverse Prob., vol. 23, pp. 969-985, Oct. 2007. https://doi.org/10.1088/0266-5611/23/3/008
  15. T. Blumensath, "Sampling and reconstructing signals from a union of linear subspaces," IEEE Trans. Inform. Theory, vol. 57, pp. 4660-4671, July 2011. https://doi.org/10.1109/TIT.2011.2146550
  16. Y. C. Eldar and G. Kutyniok, Compressed Sensing: Theory and Applications, Cambridge Univ. Press, 2012.
  17. D. L. Donoho and M. Elad, "Optimally sparse representation in general (non-orthogonal) dictionaries viaminimization," Proc. Natl. Acad. Sci. USA, vol. 100, pp. 2197-2002, Dec. 2002.
  18. C. Tomasi and R. Manduchi, "Bilateral filtering for gray and color images," Int. Conf. Computer Vision, pp. 839-846, Jan. 1998.
  19. J. E. Fowler, S. Mun, and E. W. Tramel, "Block-based compressed sensing of images and video," Foundations and Trends in Signal Process., vol. 4, pp. 297-416, Mar. 2010. https://doi.org/10.1561/2000000033
  20. K. Q. Dinh, H. Shim, and B. Jeon, "Deblocking filter for artifact reduction in distributed compressed video sensing," Visual Comm. Image Processing (VCIP), pp. 1-5, Nov. 2012.