효과적인 이미지 검색을 위한 연장 해쉬(Extendible hash) 기반 인덱싱 및 검색 기법

Indexing and Matching Scheme for Content-based Image Retrieval based on Extendible Hash

  • 탁윤식 (고려대학교 전기전자전파공학과) ;
  • 황인준 (고려대학교 전기전자전파공학과)
  • 투고 : 2010.12.03
  • 심사 : 2010.12.29
  • 발행 : 2010.12.30

초록

보다 빠른 내용 기반 이미지 검색을 위해, 다차원 특징 정보의 효과적인 인덱싱에 대한 다양한 연구들이 수행되고 있다. 하지만, 대부분의 인덱싱 기법들은 특징 정보의 차원이 커질수록 성능이 저하되는 문제를 가지고 있으며, 이를 대체하기 위해서 '높은 확률'로써 사용자가 원하는 결과를 제공해 주기 위한 휴리스틱 (heuristic) 알고리즘을 사용한 기법들이 제안되었다. 본 논문에서는 이러한 다차원 특징 정보를 효과적으로 인덱싱 하기 위해, 연장 해쉬 기반의 새로운 인덱싱 기법을 제안한다. 제안된 인덱싱 기법은 기존의 기법들이 가졌던 문제들을 해결하기 위해, 검색의 정확도에 영향을 주지 않으면서 빠른 검색이 가능하도록 설계되었다. 다양한 실험을 통해, 제안된 기법이 월등한 성능을 가질 수 있음을 보였다.

So far, many researches have been done to index high-dimensional feature values for fast content-based image retrieval. Still, many existing indexing schemes are suffering from performance degradation due to the curse of dimensionality problem. As an alternative, heuristic algorithms have been proposed to calculate the result with 'high probability' at the cost of accuracy. In this paper, we propose a new extendible hash-based indexing scheme for high-dimensional feature values. Our indexing scheme provides several advantages compared to the traditional high-dimensional index structures in terms of search performance and accuracy preservation. Through extensive experiments, we show that our proposed indexing scheme achieves outstanding performance.

키워드

참고문헌

  1. Eamonn Keogh, Li Wei, Xiaopeng Xi, Sang-Hee lee and Michail Vlachos, "LB_Keogh supports exact indexing of shapes under rotation invariance with arbitrary representations and distance measures," VLDB'06, pp.882 - 893, 2006
  2. Yoon-Sik Tak and Eenjun Hwang, "An indexing scheme for efficient camera angle invariant image retrieval," CIT'08, pp.143-148, 2008
  3. E. Keogh and C. Ratanamahatana, "Exact indexing of dynamic time warping," Knowledge and Information Systems, Vol.7, pp. 358-386, 2005 https://doi.org/10.1007/s10115-004-0154-9
  4. Shu Lin, M. Tamer Özsu, Vincent Oria and Raymond T. Ng, "An Extendible Hash for Multi-Precision Similarity Querying of Image Databases," VLDB'01, pp. 221 - 230, 2001
  5. Antonin Guttman, "R-trees: a dynamic index structure for spatial searching," Proceedings of the 1984 ACM SIGMOD international conference on Management of data, pp.47-57, 1984
  6. T. Sellis, N. Roussopoulos, and C. Faloutsos, "The R+-Tree: A dynamic index for multi-dimensional objects," In Proc. of the Int. Conference on Very Large Databases, 1987.
  7. Norbert Beckmann, Hans-Peter Kriegel, Ralf Schneider, Bernhard Seeger, "The R*-Tree: An Efficient and Robust Access Method for Points and Rectangles," SIGMOD Conference pp.322-331, 1990.
  8. D.A. White and R. Jain, "Similarity indexing with the SS-Tree," In Proc. of the 12th International Conference on Data Engineering, pp.516-523,1996
  9. N. Katayama and S. Satoh, "The SR-Tree: an index structure for high dimensional nearest neighbor queries," Iin Proc. of the ACM SIGMOD International Conference on Management of Data, pp.69-380, 1997.
  10. Yianilos and Peter N, "Data structures and algorithms for nearest neighbor search in general metric spaces," Proc. of the fourth annual ACM-SIAM Symposium on Discrete algorithms, pp. 311-321, 1993.
  11. H.Wang, C.-S.Perng, "The S2-Tree: an index structure for subsequence matching of spatial objects," Iin Fifth Pacific-Asic Conference on Knowledge Discovery and Data Mining (PAKDD), 2001.
  12. Y. Tak and E. Hwang, "A Leaf Image Retrieval Scheme Based on Partial Dynamic Time Warping and Two-Level Filtering," CIT'07, pp. 663-638, Oct. 2007