Gaussian Noise Reduction Algorithm using Self-similarity

자기 유사성을 이용한 가우시안 노이즈 제거 알고리즘

  • Published : 2007.09.25

Abstract

Most of natural images have a special property, what is called self-similarity, which is the basis of fractal image coding. Even though an image has local stationarity in several homogeneous regions, it is generally non-stationarysignal, especially in edge region. This is the main reason that poor results are induced in linear techniques. In order to overcome the difficulty we propose a non-linear technique using self-similarity in the image. In our work, an image is classified into stationary and non-stationary region with respect to sample variance. In case of stationary region, do-noising is performed as simply averaging of its neighborhoods. However, if the region is non-stationary region, stationalization is conducted as make a set of center pixels by similarity matching with respect to bMSE(block Mean Square Error). And then do-nosing is performed by Gaussian weighted averaging of center pixels of similar blocks, because the set of center pixels of similar blocks can be regarded as nearly stationary. The true image value is estimated by weighted average of the elements of the set. The experimental results show that our method has better performance and smaller variance than other methods as estimator.

대부분의 자연 영상은 프랙탈 이론의 기반이 되는 자기 유사성이라는 특징을 가지고 있다. 비록 국부적으로 영상을 정상 신호라고 가정할 수 있지만 일반적으로 영상 신호는 에지나 코너 부분과 같은 불연속성을 가지고 있는 비정상 신호이다. 이 때문에 대부분의 선형 알고리즘의 성능 저하가 나타난다. 따라서 이러한 문제를 해결하기 위하여 본 논문에서는 영상 내에 포함되어 있는 자기 유사성을 이용하는 새로운 비선영 잡음 제거 알고리즘을 제안 한다. 이를 위해 우선 잡음 제거를 수행 할 위치의 화소 주변 화소들을 이용하여 평탄 영역인지를 판단한다. 평탄 영역일 경우 그 주변 픽셀들의 평균으로 잡음을 제거하고, 평탄 영역이 아닌 경우, 블록 MSE(block Mean Square Error) 관점에서 유사도가 높은 블록을 탐색하여 그 블록들의 중심 화소값들을 이용하여 잡음 제거를 수행한다. 실험 결과는 PSNR 측면에서 잡음 제거 성능이 약 $1{\sim}3dB$ 정도 향상됨을 보여준다. 또한 추정 이론 관점에서 추정자의 분산 분석 결과 가장 낮은 분산을 갖음을 보였다.

Keywords

References

  1. Gonzalez, Woods, 'Digital Iimage Processing',Prentice Hall, 2002
  2. M. J. McDonnel, 'Box-Filtering Techniques', Computer Graphics and Image Processing pp. 65-70, 1981 https://doi.org/10.1016/S0146-664X(81)80009-3
  3. J. W. Tukey, 'Nonlinear (Nonsuperposable) Methods for Smoothing Data', in Conf. Rec., EASCON, pp.763, 1974
  4. J. S. Lee, 'Digital Image Smoothing and the Sigma Filter', Computer Graphics and Image Processing, pp.255-269, 1983
  5. Carlos A. Pomalaza-raez, Clare D. McGillem, 'An Adaptative, Nonlinear Edge-Preserving Filter', IEEE Transactions on Acoustics, Apeech, and Signal Processing, Vol. ASSP-32, No.3, June 1984
  6. Mehmet K. Ozkan, M. Ibrahim Sezan, A. Murat Tekalp, 'Adaptive Motion-Compensated filtering of Noisy Image sequences' IEEE Transactions on Circuits and Systems for Video Technology, Vol.3, No.4, August 1993
  7. C. Tomasi, R. Manduchi 'Bilateral Filtering for Gray and Color Images', IEEE International Conference on Computer Vision, Corfu, Bombay, India September 1998
  8. S. I. Olsen 'Estimation of Noise in Images : An Evaluation', Graphical Models and Image Process., Vol. 55. No. 4, pp.319-323, July 1993 https://doi.org/10.1006/cgip.1993.1022
  9. S. I. Olsen 'Estimation of Noise in Images : An Evaluation', Graphical Models and Image Process., Vol. 55. pp.319-323, July 1993 https://doi.org/10.1006/cgip.1993.1022
  10. J. B. Bednar and T. L. Watt, 'Alpha-Trimmed Means and Their Relationship to Median Filters', IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. ASSP-32, pp. 145-153, Feb. 1984
  11. Ning Lu, 'Fractal Imaging', Academic press, pp.11-13, 1997
  12. Steven M. Key 'Fundamentals of statistical signal processing', Prentice Hall, pp. 31, 1993
  13. A. K. Jain, 'Fundamentals of Digital Image Processing', Prentice-Hall, Englewood Clis, New Jersey, 1989