윤곽선 이미지 피라미드와 관심영역 검출을 이용한 SIFT 기반 이미지 유사성 검색

SIFT based Image Similarity Search using an Edge Image Pyramid and an Interesting Region Detection

  • 유승훈 (인하대학교 전자공학과) ;
  • 김덕환 (인하대학교 전자공학과) ;
  • 이석룡 (한국외국어대학교 산업경영공학부) ;
  • 정진완 (한국과학기술원 전산학과) ;
  • 김상희 (국방과학연구소 지형영상처리팀)
  • 발행 : 2008.08.15

초록

다양한 형태 특징 추출 방법 중의 하나인 SIFT는 물체 인식, 모션 추적, 3차원 이미지 재구성과 같은 컴퓨터 비전 응용 분야에서 많이 사용된다. 하지만 SIFT 방법은 많은 특징점들과 고차원의 특징 벡터를 사용하기 때문에 이미지 유사성 검색에 그대로 적용하기에는 많은 어려움이 있다. 본 논문에서는 윤곽선 이미지 피라미드와 관심영역 검출을 이용한 SIFT 기반 이미지 유사성 검색 기법을 제안한다. 제안한 방법은 윤곽선 이미지 피라미드를 이용하여 이미지의 밝기 변화, 크기, 회전등에 불변한 특징을 추출하고, 타원 형태의 허프변환을 이용한 관심영역 검출을 통해 불필요한 많은 특징점들을 제거하여 검색성능을 높인다. 실험 결과에서 제안한 방법의 이미지 검색 성능이 기존의 SIFT의 방법에 비해 평균 재현율이 약 20%정도 좋은 성능을 보이고 있다.

SIFT is popularly used in computer vision application such as object recognition, motion tracking, and 3D reconstruction among various shape descriptors. However, it is not easy to apply SIFT into the image similarity search as it is since it uses many high dimensional keypoint vectors. In this paper, we present a SIFT based image similarity search method using an edge image pyramid and an interesting region detection. The proposed method extracts keypoints, which is invariant to contrast, scale, and rotation of image, by using the edge image pyramid and removes many unnecessary keypoints from the image by using the hough transform. The proposed hough transform can detect objects of ellipse type so that it can be used to find interesting regions. Experimental results demonstrate that the retrieval performance of the proposed method is about 20% better than that of traditional SIFT in average recall.

키워드

참고문헌

  1. D. Lowe, "Object Recognition from Local Scale- Invariant Features," In International Conference on Computer Vision, pp. 1150-1157, 1999
  2. D. Lowe, "Distinctive Image Features from Scale- Invariant Keypoints," Int'l J. Computer Vision, Vol.2, No.60, pp. 91-110, 2004
  3. 한재선, 이광국, 김회율, "SIFT와 ART를 이용한 지역서술자," 제 19회 신호처리합동학술대회, Vol.19, 2006
  4. K Mikolajczyk, C Schmid, "An affine invariant interest point detector," ECCV, 2002
  5. Y Ke, R Sukthankar, "PCA-SIFT: A more distinctive representation for local image descriptors," IEEE CVPR, 2004
  6. W Cheung, G Hamarneh, "N-SIFT: N-DIMENSIONAL SCALE INVARIANT FEATURE TRANSFORM FOR MATCHING MEDICAL IMAGES," IEEE International Symposium on Biomedical Imaging, 2007
  7. C. Harris, M.J. Stephens, "A combined corner and edge detector," In Alvey Vision Conference, pp. 147-152, 1988
  8. H Bay, T Tuytelaars, L Van Gool, "SURF: Speeded Up Robust Features," ECCV, 2006
  9. Yan Ke, Rahul Sukthankar, Larry Huston, "An efficient parts-based near-duplicate and sub-image retrieval system," Proceedings of the 12th annual ACM international conference on Multimedia, October 10-16, 2004
  10. J. J. Foo, J. Zobel, R. Sinha, and S. Tahaghoghi, "Detection of near-duplicate images for web search," In Proc. Conference On Image And Video Retrieval, July 2007
  11. Duda, R. O. and P. E. Hart, "Use of the Hough Transformation to Detect Lines and Curves in Pictures," Comm. ACM, Vol.15, pp. 11-15, 1972 https://doi.org/10.1145/361237.361242
  12. R.Fergus, P.Perona, A.Zisserman, "Object Class Recognition by Unsupervised Scale-Invariant Learning," IEEE CVPR, 2003
  13. Pawel Janusz Czerwi ski, Ryad Chellali, Vincent Fremont, "Ellipse Detection Using Hough Transform," Warsaw University of Technology and Ecole des Mines de Nantes, 2003
  14. R. C. Gonzalez and R. E.Woods, "Digital Image Processing 3rd edition," Addison-Wesley, 2007
  15. Saburo Tsuji, Fumio Matsumoto, "Detection of Ellipses by a Modified Hough Transformation," IEEE Transactions on Computers, 1978
  16. Dimitrios Ioannoua, Walter Hudab, Andrew F. Laine, "Circle recognition through a 2D Hough Transform and radius histogramming," Image and Vision Computing, 1999
  17. L Xu, E Oja, "Randomized Hough transform (RHT): basic mechanisms, algorithms, and computational complexities," CVGIP: Image Understanding, 1993
  18. John Canny, "A Computational Approach to Edge Detection," IEEE Transactions on PAMI, 1986
  19. http://staff.science.uva.nl/~aloi/