DOI QR코드

DOI QR Code

퍼지 유사성 기반 슈퍼-픽셀 생성

Super-Pixels Generation based on Fuzzy Similarity

  • 김용길 (조선이공대학교 컴퓨터보안과) ;
  • 문경일 (호남대학교 공과대학 컴퓨터공학과)
  • 투고 : 2017.01.15
  • 심사 : 2017.04.07
  • 발행 : 2017.04.30

초록

최근에는 슈퍼-픽셀 (super-pixel)은 컴퓨터 발전 응용에 널리 사용되고 있다. 슈퍼 픽셀 알고리즘은 픽셀을 지각적으로 실행이 가능한 영역으로 변환하여 그리드 픽셀의 경직된 특징을 줄일 수 있다. 특히, 슈퍼 픽셀은 깊이 추정, 골격 작업, 바디 라벨링 및 기능 국소화 등에 사용된다. 그러나 이러한 작업을 수행하기 위해 우수한 슈퍼 픽셀 파티션을 생성하는 것은 쉽지 않다. 특히 슈퍼 픽셀은 비합, 지속, 폐쇄, 지각 불변과 같은 형태 측면을 고려할 때보다 의미있는 특징을 만족시키지는 못한다. 본 논문에서는 단순 선형 반복 클러스터링과 퍼지 클러스터링 개념을 결합한 고급 알고리즘을 제안한다. 단순 선형 반복 클러스터링 기술은 이미지 경계, 속도, 메모리 효율이 기존 방법보다 높다. 그것은 형태 측면의 맥락에서 슈퍼 픽셀 형태에 대해 양호하게 작거나 규칙적인 특성을 제안하는 것은 아니다. 퍼지 유사성 측정은 제한된 크기와 이웃을 고려하여 합리적인 그래프를 제공한다. 보다 작고 규칙적인 픽셀을 얻으며 부분적으로 관련된 특징을 추출 할 수 있다. 시뮬레이션은 퍼지 유사성 기반 슈퍼 픽셀 생성은 사람의 이미지를 분해하는 방식으로 자연적 특징을 대표적으로 나타낸다.

In recent years, Super-pixels have become very popular for use in computer vision applications. Super-pixel algorithm transforms pixels into perceptually feasible regions to reduce stiff features of grid pixel. In particular, super-pixels are useful to depth estimation, skeleton works, body labeling, and feature localization, etc. But, it is not easy to generate a good super-pixel partition for doing these tasks. Especially, super-pixels do not satisfy more meaningful features in view of the gestalt aspects such as non-sum, continuation, closure, perceptual constancy. In this paper, we suggest an advanced algorithm which combines simple linear iterative clustering with fuzzy clustering concepts. Simple linear iterative clustering technique has high adherence to image boundaries, speed, memory efficient than conventional methods. But, it does not suggest good compact and regular property to the super-pixel shapes in context of gestalt aspects. Fuzzy similarity measures provide a reasonable graph in view of bounded size and few neighbors. Thus, more compact and regular pixels are obtained, and can extract locally relevant features. Simulation shows that fuzzy similarity based super-pixel building represents natural features as the manner in which humans decompose images.

키워드

참고문헌

  1. R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua and S. Susstrunk, "SLIC Super pixels Compared to State-of-the-Art Super pixel Methods," IEEE Transactions on pattern analysis and machine intelligence, Vol. 34, No. 11. 2012, pp. 2274-2282. https://doi.org/10.1109/TPAMI.2012.120
  2. A. Ayvaci and S. Soatto, "Motion segmentation with occlusions on the super pixel graph," Proceedings of the Workshop on Dynamical Vision, Kyoto, Japan, 2009.
  3. A. V. Baterina, "Image Edge Detection Using Ant Colony Optimization", International Journal of Circuits, Systems and Signal Processing, Issue 2, Vol. 4. No. 2. 2010, pp.58-67.
  4. D. Comaniciu and P. Meer, "Mean shift: a robust approach toward feature space analysis." Pattern Analysis and Machine, Vol. 24, No. 5, 2002, pp. 603-619. https://doi.org/10.1109/34.1000236
  5. P. Felzenszwalb and D. Huttenlocher, "Efficient graph-based image segmentation." IJCV, 2004, pp. 167-181.
  6. B. Fulkerson, A. Vedaldi, S. Soatto, "Class segmentation and object localization with super-pixel neighborhoods." Computer Vision, IEEE 12. 2009.
  7. X. He, R. Zemel, D. Ray, "Learning and incorporating top-down cues in image segmentation." ECCV, 2006, pp. 338-351.
  8. D. Hoiem, A. Efros, M. Hebert, "Automatic Photo Pop-up." ACM Transactions on Graphics, Vol. 24, 2005, pp. 577-584. https://doi.org/10.1145/1073204.1073232
  9. T. Kanungo, D. M. Mount, N. S. Netanyahu, C.D. Piatko, R. Silverman and A. Y. Wu, "A local search approximation algorithm for k-means clustering." Symposium on Computational geometry, 2002, pp.10-18.
  10. M. Jafar, "Ant-based Clustering Algorithms: A Brief Survey", International Journal of Computer Theory and Engineering, Vol. 2, No. 5, 2010, pp. 1793-8201.
  11. A. Levinshtein, C. Sminchisescu and S. Dickinson, "Multi-scale symmetric part detection and grouping." International Journal of Computer Vision September, Vol. 104, No. 2, 2013, pp.117-134. https://doi.org/10.1007/s11263-013-0614-3
  12. A. Levinshtein, A. Stere, K. Kutulakos, D. Fleet, S. Dickinson and K. Siddiqi, "Turbo pixels: Fast super pixels using geometric flows." IEEE Transactions on Pattern Analysis & Machine Intelligence 2009, Vol.31, No.12. 2009, pp. 2290- 2297.
  13. Y. Li, J. Sun, C. K. Tang, H. Y. Shum, "Lazy snapping." ACM Transactions on Graphics, Vol. 23, 2004, pp. 303-308. https://doi.org/10.1145/1015706.1015719
  14. A. Moore, S. Prince, J. Warrell, U. Mohammed and G. Jones, "Super pixel Lattices." IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2008), 2008, pp. 24-26.
  15. G. Mori, "Guiding model search using segmentation." Computer Vision, IEEE, 2005, pp.1417-1423.
  16. X. Ren and J. Malik, "Learning a classification model for segmentation." International Conference on Computer Vision - ICCV, 2003, pp. 10-17.
  17. J. Shi and J. Malik, "Normalized cuts and image segmentation." PAMI, 2000, pp. 888-905
  18. A. Vedaldi and S. Soatto, "Quick shift and kernel methods for mode seeking." Computer Vision, ECCV , 2008, pp.705-718.
  19. L. Vincent and P. Soille, "Watersheds in digital spaces: An efficient algorithm based on immersion simulations." PAMI, 1991, pp.583-598.
  20. S.H. Lee, H. Jeong and K.I. Moon, "Image Segmentation Based on Fuzzy Clustering Super-Pixel," Proc. Advanced and Applied Convergence Letters, Yanbian, China, June 30-july 4, AACL 05, 2015, pp. 177-180.
  21. B. Sridevi and R. Nadarajan, "Fuzzy similarity measure for generalized fuzzy numbers," Int. J. Open Problems Compt. Math., Vol. 2, No. 2, June 2009, pp.240-253.
  22. S. Gould, J. Rodgers, D. Cohen, G. Elidan, and D. Koller, "Multi-class segmentation with relative location prior," Int'l J. Computer Vision, vol. 80, no. 3, 2008, pp. 300-316. https://doi.org/10.1007/s11263-008-0140-x