Korean Journal of Remote Sensing (대한원격탐사학회지)

Korean Society of Remote Sensing (대한원격탐사학회)
권호 표지
DOI QR코드

DOI QR Code

RAG-based Image Segmentation Using Multiple Windows

RAG 기반 다중 창 영상 분할 (1)

  • 이상훈 (경원대학교 산업공학과)
  • Published : 2006.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes RAG (Region Adjancency Graph)-based image segmentation for large imagery in remote sensing. The proposed algorithm uses CN-chain linking for computational efficiency and multi-window operation of sliding structure for memory efficiency. Region-merging due to RAG is a process to find an edge of the best merge and update the graph according to the merge. The CN-chain linking constructs a chain of the closest neighbors and finds the edge for merging two adjacent regions. It makes the computation time increase as much as an exact multiple in the increasement of image size. An RNV (Regional Neighbor Vector) is used to update the RAG according to the change in image configuration due to merging at each step. The analysis of large images requires an enormous amount of computational memory. The proposed sliding multi-window operation with horizontal structure considerably the memory capacity required for the analysis and then make it possible to apply the RAG-based segmentation for very large images. In this study, the proposed algorithm has been extensively evaluated using simulated images and the results have shown its potentiality for the application of remotely-sensed imagery.

본 연구는 방대한 크기 원격 탐사 영상 자료의 효율적인 분석을 위한 RAG (Region Adjancency Graph) 기반 영상 분할 기법을 제안하고 있다. 제안된 알고리즘은 계산의 효율성을 위하여 CN-chain 연결과 저장 기억의 효율성을 위하여 sliding 다중 창을 사용한다. RAG에 의한 지역 합병은 최선의 결합을 위한 edge을 발견과 합병에 따른 graph의 갱신의 과정이다. CN-chain 연결법은 가장 유사한 인접 지역의 연결을 형성하면서 최선의 edge를 발견하여 합병을 해 나가는 과정으로 영상 자료 크기의 증가에 따라 단지 증가 배수만큼만 분석 시간을 증가시킨다. 합병에 의해 변하는 RAG의 효율적인 갱신을 위하여 RNV(Regional Neighbor Vector)를 사용하였다. 방대한 크기 자료 분석은 막대한 기억 용량의 시스템을 필요로 한다. 제안된 수평적인 구조의 sliding 다중 창 작업은 필요한 기억 용량을 현저히 감소시켜 가능한 분석 자료의 크기를 증대시켰을 뿐 아니라 계산 시간의 감소를 초래하였다. 본 연구는 simulation 자료를 사용하여 광범위하게 제안된 알고리즘의 성능을 실험하였으며 실험 결과는 알고리즘의 효율성을 입증하였다.

Keywords

References

  1. 이상훈, 2001. 공간지역확장과 계층집단연결 기법을 이용한 무감독 영상분류, 대한원격탐사학회지, 17: 57-70
  2. 이상훈, 2006, RAG기반 계층 분류, 대한원격탐사학회지, 22권 6호에 게재 예정 https://doi.org/10.7780/kjrs.2006.22.6.613
  3. Adams, R. and L. Bischof, 1994. Seeded region growing, IEEE Trans. Pattern Anal. Machine Intell., 16: 641-647 https://doi.org/10.1109/34.295913
  4. Anderberg, M. R., 1973. Cluster Analysis for Application, Academic Press, NY
  5. Ballard, D. and C. Brown, 1982. Computer Vision. Englewood Cliffs, NJ: Prentice-Hall
  6. Canny, J., 1986. A computational approach to edge detection, IEEE Trans. Pattern Anal. Machine. Intell., 8: 679-698 https://doi.org/10.1109/TPAMI.1986.4767851
  7. Chang, Y-L. and X. Li, 1994. Adaptive image region growing, IEEE Trans. Image Process., 3: 868-872 https://doi.org/10.1109/83.336259
  8. Chen, S, W. Lin, and C. Chen, 1991. Split-and-merge image segmentation based on localized feature analysis and statistical tests, CVGIP: Graph. Models Image Process., 53: 457-475 https://doi.org/10.1016/1049-9652(91)90030-N
  9. Fan, J., D. K. Y. Yau, A. K. Elmagarmid, and W. G. Aref, 2001. Automatic image segmentation by integrating color-edge extraction and seeded region growing, IEEE Trans. Image Process., 10: 1454-1466 https://doi.org/10.1109/83.951532
  10. Haralick, R. and L. Shapiro, 1985. Image segmentation techniques, CVGIP, 29: 100-132
  11. Haris, K., S. N. Efstratiadis, N. Maglaveras, and A. K. Katsaggelos, 1998. Hybrid image segmentation usingwatershed and fast region merging, IEEE Trans. Image Process., 7: 1684-1699 https://doi.org/10.1109/83.730380
  12. Hojjatoleslami, S. A. and J. Kittler, 1998. Region growing: a new approach, IEEE Trans. Image Process., 7: 1079-1084 https://doi.org/10.1109/83.701170
  13. Jain, A., 1989. Fundamentals of Digital Image Processing, Englewood Cliffs, NJ; Prentice-Hall
  14. Lee, S., 1989. An unsupervised hierarchical clustering image segmentation and an adaptive image reconstruction system for remote sensing, Ph.D. Thesis, The University of Texas at Austin
  15. Lee, S-H, 2004. Unsupervised image classification using region-growingsegmentation based on CN-chain, Korean Journal of Remote Sensing, 20: 215-225 https://doi.org/10.7780/kjrs.2004.20.3.215
  16. Lee, S. and M. M. Crawford, 2005. Unsupervised Bayesian image segmentation using multistage hierarchical clustering, IEEE Trans. Image Process., 14: 312-320 https://doi.org/10.1109/TIP.2004.841195
  17. Pal, N. and S. Pal, 1993. A review on image segmentation techniques, Pattern Recognit., 26: 1277-1294 https://doi.org/10.1016/0031-3203(93)90135-J
  18. Pavlidis, T. and Y-T. Liow, 1990. Integrating region growing and edge detection, IEEE Trans. Pattern Anal. Machine Intell., 12: 225-233 https://doi.org/10.1109/34.49050
  19. Sahoo, P. K., S.Soltani, A. K. C. Wong, and Y. C. Chen, 1988. A survey of thresholding techniques, CVGIP, 41: 233-260
  20. Schwarz, G., 1978. Estimation of the Dimension of a Model, Annal., Math., Statist., 6: 461-464 https://doi.org/10.1214/aos/1176344136
  21. Tanimoto, S. and A. Klinger, 1980. Structured Computer Vision, Academic, NY
  22. Tobias, O. J. and R. Seara, 2002. Image segmentation by histogram thresholding using fuzzy sets, IEEE Trans. Image Process., 11: 1457-1465 https://doi.org/10.1109/TIP.2002.806231
  23. Torre V. and T. Poggio, 1986. On edge detection, IEEE Trans. Pattern Anal. Machine Intell., 8: 147-163 https://doi.org/10.1109/TPAMI.1986.4767769
  24. Wan, S.-Y and W. E. Higgins, 2003. Symmetric region growing, IEEE Trans. Image Process. 12: 1007-1015 https://doi.org/10.1109/TIP.2003.815258
  25. Won, C. S. and H. Derin, 1992. Unsupervised segmentation of noisy and textured images using Markov randomfields, CVGIP, 54: 308-328
  26. Wu, Z., 1993. Homogeneity testing for unlabeled data: A performance evaluation, CVGIP: Graph. Models Image Process., 55: 370-380 https://doi.org/10.1006/cgip.1993.1028
  27. Zhu S. C. and A. Yuille, 1996. Region growing, and Bayes/MDL for multiband image segmentation, IEEE Trans. Pattern Anal. Machine Intell., 18: 884-900 https://doi.org/10.1109/34.537343