The Transactions of the Korea Information Processing Society (한국정보처리학회논문지)

Korea Information Processing Society (한국정보처리학회)
권호 표지

Unsuperised Image Segmentation Algorithm Using Markov Random Fields

마르코프 랜덤필드를 이용한 무관리형 화상분할 알고리즘

  • 박재현 (명지대학교 전자정보통신공학부)
  • Published : 2000.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, a new unsupervised image segmentation algorithm is proposed. To model the contextual information presented in images, the characteristics of the Markov random fields (MRF) are utilized. Textured images are modeled as realizations of the stationary Gaussian MRF on a two-dimensional square lattice using the conditional autoregressive (CAR) equations with a second-order noncausal neighborhood. To detect boundaries, hypothesis tests over two masked areas are performed. Under the hypothesis, masked areas are assumed to belong to the same class of textures and CAR equation parameters are estimated in a minimum-mean-square-error (MMSE) sense. If the hypothesis is rejected, a measure of dissimilarity between two areas is accumulated on the rejected area. This approach produces potential edge maps. Using these maps, boundary detection can be performed, which resulting no micro edges. The performance of the proposed algorithm is evaluated by some experiments using real images as weB as synthetic ones. The experiments demonstrate that the proposed algorithm can produce satisfactorY segmentation without any a priori information.

본 논문에서는 새로운 무관리형 화상분할 알고리즘이 제안된다. 제안된 알고리즘은 화상에 내재되어 있는 구조 정보를 모델링하기 위하여 마르코프 랜덤필드의 특성을 이용하고 있다. 텍스쳐 화상은 정상상태의 가우스 마르코프 랜덤필드가 2차원의 격자구조 위에 실현된 상태로 간주되었으며 2차의 비순차근방을 갖는 조건부 자기회귀함수를 이용하여 모델링 되었다. 화상의 경계면 감출을 위하여 마스크로 선택된 두 영역에 대한 가설검정이 수행된다. 이 방법은 선택된 두 영역이 같은 종류의 텍스쳐라고 가정을 한 후 조건부 자기회귀모델의 매개변수를 최소평균제곱오차 측면에서 추정한다. 가설이 거절되면 두 영역의 상이함을 측정한 그 값이 선택된 영역에 누적된다. 이와 겉은 방법을 통하여 잠재적인 경제지도가 얻어지며, 이것을 통하여 여러 종류의 텍스쳐 화상의 분할이 미세오류경계 없이 이루어지게 된다. 제안된 알고리즘의 성능은 인공화상 뿐만 아니라 실제의 자연화상을 이용한 실험을 통하여 입증되었으며 일체의 사전정보 없이도 만족할 만한 결과를 보여 주었다.

Keywords

References

  1. J. M. Francos, A. Z. Meiri, and B. Porat, 'A unified texture model based on a 2-D Wold-like decomposition,' IEEE Trans. Signal Proc. Vol.41, No.8, pp.2665-2678, Aug. 1993 https://doi.org/10.1109/78.229897
  2. A. K. Jain, 'Fundamentals of Digital Image Processing', Prentice-Hall, Englewood Cliffs, NJ, 1989
  3. R. M. Haralick, 'Statistical and structural approaches to texture,' Proc. IEEE, Vol.67, May 1979
  4. G. G. Lendaris and G. L. Stanley, 'Diffraction pattern sampling for automatic pattern recognition,' in Computer Methods in Image Analysis, Los Angeles : IEEE Computer Society, 1977
  5. R. P. Kruger, W. B. Thompson, and A. F. Turner, 'Computer diagnosis of Pneumoconiosis,' IEEE Tran. Syst., Man, Cybern. Vol.SMC-4, pp.40-49, Jan. 1974
  6. D. B. Cooper, H. Elliot, F. S. Cohen, and P. F. Symosek, 'Stochastic boundary estimation and object recognition,' Computer Graphics and Image Processing, pp.326-355, Apr. 1980 https://doi.org/10.1016/0146-664X(80)90018-0
  7. F. S. Cohen and D. B. Cooper, 'Real-time texture image segmentation based on noncausal Markovian random field models,' Proc. SPIE, Vol.449, pp.17-28, 1984
  8. M. P. Dubuisoon Jolly and A. Gupta, 'Color and Texture fusion : application to aerial image segmentation and GIS updating,' Image and Vision Computing, Vol.18, Issue 10, pp.823-832, July 2000 https://doi.org/10.1016/S0262-8856(99)00050-5
  9. Z. Fan and F. Cohen, 'Textured image segmentation as a multiple hypothesis test,' IEEE Trans. Circuits Syst. Vol.35, pp.691-701, June 1988 https://doi.org/10.1109/31.1806
  10. H. Derin and H. Elliott, 'Modeling and segmentation of noisy and textured images using Gibbs random fields,' IEEE Trans. Pattern Anal. Machine Intell. Vol.PAMI-9, pp.39-55, Jan. 1987
  11. H. Elliot, H. Derin, R. Cristi, and D. Geman, 'Application of the Gibbs distribution to image segmentation,' in Proc. 1984 Int. Conf. Acoust., Speech, Signal Processing, ICASSP'84, San Diego, CA, pp.32.5.1-32.5.4, Mar. 1984
  12. F. S. Cohen and D. B. Cooper, 'Simple parallel hierarchical and relaxation algorithms for segmenting noncausal Markovian random fields,' IEEE Trans. Pattern Anal. Machine Intell. Vol.PAMI-9, No.2, pp.195-219, Mar. 1987
  13. S. Geman and C. Geman, 'Stochastic relaxation, Gibbs distribution, and the Bayesian restoration of images,' IEEE Trans. Pattern Anal. Machine Intell. Vol.PAMI-6, pp.721-741. Nov. 1984
  14. D. Stern and L. Kurz, 'Edge detection in correlated noise using Latin square masks,' Pattern Recognition, Vol.2, No.2, pp.119-129, 1988
  15. M. M. Benteftifa and L. Kurz., 'Two-dimensional object detection in correlated noise,' Pattern Recognition, Vol.24, pp.755-774, 1991 https://doi.org/10.1016/0031-3203(91)90044-6
  16. S. H. Chang and L. Kurz, 'Object detection and experimental designs,' Computer Graphics, Vision, and Image Processing, Vol.40, pp.147-168, 1987 https://doi.org/10.1016/S0734-189X(87)80112-3
  17. B. Julesz, 'Experiments in the visual perception of texture,' Scientific American, Vol.232, pp.765-771, June 1977
  18. J. E. Besag, 'Spatial interaction and the statistical analysis of lattice systems,' J. Royal Stat. Soc., Ser. B, Vol.B-36, pp.192-236, 1974
  19. J. W. Woods, 'Two-dimensional discrete Markovian fields,' IEEE Trans : Inform. Theory, Vol.IT-18, pp.232-240, Mar. 1972 https://doi.org/10.1109/TIT.1972.1054786
  20. C. R. Rao, 'Linear Statistical Inference and Its Applications', 2nd ed., Wiley, New York, 1973
  21. Lei Wang and Jun Liu, 'Texture Segmentation based on MRMRF modeling,' Pattern Recognition Letters, Vol.21, Issue 2, pp.189-200, Feb. 2000 https://doi.org/10.1016/S0167-8655(99)00146-4