다중해상도 탐색을 이용한 반복 일반화 허프 변환

Iterative Generalized Hough Transform using Multiresolution Search

  • 이경미 (덕성여자대학교 컴퓨터과학부) ;
  • ;
  • W. Nick Street (Department of Management Sciences at the University)
  • 발행 : 2003.10.01

초록

이 논문은 주어진 영상에 존재하는 물체를 자동적으로 탐지하기 위한 시간과 공간 효율적인 방법을 소개한다. 일반화 허프 변환(Generalized Hough Transform: GHT)은 다양한 모양의 물체를 찾기 위해 자동 물체 탐지를 하는 강력한 템플릿(template) 매칭 알고리즘이다. 다양한 모양과 크기의 물체를 찾기 위해 서로 다른 많은 템플릿을 GHT에 적용해야 한다. GHT로 찾아진 모든 경계선은 보다 정교한 경계선을 찾기 위한 초기 외곽선으로 사용된다. 그러나, GHT의 주요 단점은 과도한 시간과 공간을 요구하는 것이다. 이런 단점을 극복하기 위해서, 제안된 알고리즘은 공간 효율적 방법인 반복적 GHT(iterative GHT: IGHT)를 사용한다. 또한, 원래 영상의 크기를 이분의 일 크기와 사분의 일 크기로 줄여서 다중 해상도 탐색을 이용한다. 사분의 일 영상에서 첫 번째 IGHT를 수행하여 획득한 정보를 이용하고, 세포 크기의 범위를 줄여 이분의 일 크기의 영상에서 탐색공간을 제한한다. 이분의 일 크기의 영상에서 두 번째 IGHT를 수행한 후, 세포핵은 세부 탐색에 의해 찾아지고, 정확한 경계선을 결정하기 위한 에지 정보에 의해 분할된다. 실험결과는 이 방법이 정확도의 손실이 없으면서, 수행시간과 메모리 사용을 줄이고 있음을 보여준다.

This paper presents an efficient method for automatically detecting objects in a given image. The GHT is a robust template matching algorithm for automatic object detection in order to find objects of various shapes. Many different templates are applied by the GHT in order to find objects of various shapes and size. Every boundary detected by the GHT scan be used as an initial outline for more precise contour-finding techniques. The main weakness of the GHT is the excessive time and memory requirements. In order to overcome this drawback, the proposed algorithm uses a multiresolution search by scaling down the original image to half-sized and quarter-sized images. Using the information from the first iterative GHT on a quarter-sized image, the range of nuclear sizes is determined to limit the parameter space of the half-sized image. After the second iterative GHT on the half-sized image, nuclei are detected by the fine search and segmented with edge information which helps determine the exact boundary. The experimental results show that this method gives reduction in computation time and memory usage without loss of accuracy.

키워드

참고문헌

  1. D. H. Ballard, 'Generalizing the Hough transform to detect arbitrary shapes,' Pattern Recognition, Vol. 13, No. 2, pp. 111-122, 1981 https://doi.org/10.1016/0031-3203(81)90009-1
  2. K.-M. Lee and W. N. Street, 'A fast and robust approach for automated segmentation of breast cancer nuclei,' Proceedings of the IASTED International Conference on Computer Graphics and Imaging, pp. 42-47, 1999
  3. National Alliance of Breast Cancer Organizations (NABCO). Facts about breast cancer in the USA. New York, NY, USA, February 2001. http://www.nabco.org/resources/facts/usafacts.html
  4. T. Mouroutis, S. J. Roberts, and A. A. Bharath, 'Robust cell nuclei segmentation using statistical modelling,' IOP Bioimaging, Vol. 6, No. 2, pp. 79-91, 1998 10.1002/1361-6374(199806)6:2<79::AID-BIO3>3.0.CO;2-#
  5. W. H. Wolberg, W. N. Street, and O. L. Mangasarian, 'Machine learning techniques to diagnose breast cancer from image-processed nuclear features of fine needle aspirates,' Cancer Lerrers, Vol. 77, pp. 163-171, 1994 10.1016/0304-3835(94)90099-X
  6. O. L. Mangasarian, W. N. Street, and W. H. Wolberg, 'Breast cancer diagnosis and prognosis via linear programming,' Operations Research, Vol. 43, No. 4, pp. 570-576, 1995
  7. W. N. Street, 'Xcyt: A system for remote cytological diagnosis and prognosis of breast cancer,' In L. C. Jain, editor, Soft Computing Techniques in Breast Cancer Prognosis and Diagnosis, pp. 297-322. World Scientific Publishing, 2000
  8. A. S. Aguado and M. S. Nixon. A new Hough transform mapping for ellipse detection. Technical report, University of Southampton, UK, 1995. 1995/6 Research Journal Image, Speech and Intelligent Systems
  9. A. A. Kassim, T. Tan, and K. H. Tan. 'A comparative study of efficient generalised Hough transform techniques,' Image and Vision Computing, Vol. 17, No. 10, pp. 737-748, 1999 10.1016/S0262-8856(98)00156-5
  10. D. Ma and X. Chen, 'Hough transform using slope and curvature as local properties to detect arbitrary 2D shapes,' In Proceedings of the 9th International Conference on Pattern Recognition, pp. 511-513, 1988 https://doi.org/10.1109/ICPR.1988.28280
  11. S.-C. Jeng and W.-H. Tsai, 'Scale and orientation-invariant generalized Hough transform a new approach,' Pattern Recognition, Vol. 24, No. 11, pp. 1037-1051, 1991 https://doi.org/10.1016/0031-3203(91)90120-T
  12. M. Lee, J. Kittler, and K. C. Wong, 'Generalized Hough transform in object recognition,' In Proceedings of the International Conference on Pattern Recognition, Vol. 3, pp. 285-289, 1992 Lee, H.M.;Kittler, J.;Wong, K.C. https://doi.org/10.1109/ICPR.1992.201981
  13. K.-M. Lee and W. N. Street, 'Model-based detection, segmentation and classification for image analysis using on-line shape learning,' Machine Vision and Applications, Vol. 13, No. 4, 222-233, 2003 https://doi.org/10.1007/s00138-002-0061-6
  14. H. A. Rowley, S. Baluja, and T. Kanade, 'Neural network-based face detection,' IEEE Transacations on Pattern Analysis and Machine Intelligence, Vol. 20, No. 1, pp. 23-38, 1998 https://doi.org/10.1109/34.655647
  15. H. Li, M. A. Lavin, and R. J. LeMaster, 'Fast Hough transform: A hierarchical approach,' Computer Vision, Graphics, and Image Processing, Vol. 36, No. 2-3, pp. 139-161, 1986 10.1016/0734-189X(86)90073-3
  16. V. Chalana, W. Costa, and Y. Kim, 'Integrating region and edge information using regularization,' In Proceedings of the SPIE Conference on Medical Imaging, Vol. 2434, pp. 262-271, 1995 https://doi.org/10.1117/12.208697
  17. M. Kass, A. Witkin, and D. Terzopoulos, 'Snakes: Active contour models,' International Journal of Computer Vision, Vol. 1, No. 4, pp. 321-331, 1988 https://doi.org/10.1007/BF00133570
  18. K.-M. Lee and W. N. Street, 'Incremental Feature Weight Learning and its application to Shapebased Query System,' Pattern Recognition Letters, Vol. 23, No. 7, 865-874, 2002 10.1016/S0167-8655(01)00161-1
  19. Pattern Recoginition Letters v.23 no.7 Incremental Feature Weight Learning and its application to Shapebased Query System K.M.Lee;W.N.Street