The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지
DOI QR코드

DOI QR Code

A Fast and Efficient Haar-Like Feature Selection Algorithm for Object Detection

객체검출을 위한 빠르고 효율적인 Haar-Like 피쳐 선택 알고리즘

  • 정병우 (서강대학교 전자공학과 CAD&ES 연구실) ;
  • 박기영 (서강대학교 전자공학과 CAD&ES 연구실) ;
  • 황선영 (서강대학교 전자공학과 CAD&ES 연구실)
  • Received : 2013.04.30
  • Accepted : 2013.06.05
  • Published : 2013.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

This paper proposes a fast and efficient Haar-like feature selection algorithm for training classifier used in object detection. Many features selected by Haar-like feature selection algorithm and existing AdaBoost algorithm are either similar in shape or overlapping due to considering only feature's error rate. The proposed algorithm calculates similarity of features by their shape and distance between features. Fast and efficient feature selection is made possible by removing selected features and features with high similarity from feature set. FERET face database is used to compare performance of classifiers trained by previous algorithm and proposed algorithm. Experimental results show improved performance comparing classifier trained by proposed method to classifier trained by previous method. When classifier is trained to show same performance, proposed method shows 20% reduction of features used in classification.

본 논문은 객체검출(object detection)에 사용되는 분류기의 학습을 위한 빠르고 효율적인 Haar-like feature 선택 알고리듬을 제안한다. 기존 AdaBoost를 이용한 Haar-like feature 선택 알고리듬은 학습 샘플들에 대한 피쳐의 에러만을 고려하여 형태적으로 유사하거나 중복되는 피쳐가 선택되는 경우가 많았다. 제안하는 알고리듬은 피쳐의 형태와 피쳐간의 거리로부터 피쳐의 유사도를 계산하고 이미 선택된 피쳐와 유사도가 큰 피쳐들을 피쳐 세트에서 제거하여 빠르고 효율적인 피쳐 선택이 이루어지도록 하였다. FERET 얼굴 데이터베이스를 사용하여 제안된 알고리듬을 사용하여 학습시킨 분류기와 기존 알고리듬을 사용한 분류기의 성능을 비교하였다. 실험 결과 제안한 피쳐 선택 방법을 사용하여 학습시킨 분류기가 기존 방법을 사용한 분류기보다 향상된 성능을 보였으며, 동일한 성능을 갖도록 학습시켰을 경우 분류기의 피쳐 수가 20% 감소하였다.

Keywords

References

  1. J. Lee, J, Ryu, S. Hong, and W. Cho, "Local and global collaboration for object detection enhancement with information redundancy," in Proc. 6th IEEE Int. Conf. Adv. Video and Signal Based Surveillance, pp. 358-363, Genova, Italy, Sep. 2009.
  2. M. Yang, D. Kriegman, and N. Ahuja, "Detecting faces in images: a survey," IEEE Trans. Pattern Anal. Mach. Intell., vol. 24, no. 1, pp. 34-58, Jan. 2002. https://doi.org/10.1109/34.982883
  3. P. Viola and M. Jones, "Robust real-time face detection," in Proc. 8th IEEE Int. Conf. Comput. Vision, vol. 2, pp. 747, Vancouver, Canada, July 2001.
  4. Y. Freund and R. Schapire, "A decision-theoretic generalization of on-line learning and an application to boosting," Computational Learning Theory: J. Comput. Syst. Sci., vol. 55, no. 1, pp. 119-139, Aug. 1997.
  5. J. Ren, N. Kehtarnavaz, and L. Estevez, "Real-time optimization of Viola-Jones face detection for mobile platforms," in Proc. 7th IEEE Dallas Circuits and Syst. Workshop, vol. 1, no.1, pp. 1-4, Dallas, U.S.A., Oct. 2008.
  6. M. Kolsch and M. Turk, "Analysis of rotational robustness of hand detection with a Viola-Jones detector," in Proc. 17th Int. Conf. Pattern Recognition, vol. 3, pp. 107-110. Cambridge, U.K., Aug. 2004.
  7. L. Acasandrei and A. Barriga-Barros, "Accelerating Viola-Jones face detection for embedded and SoC environments," in Proc. 5th ACM/IEEE Int. Conf. Distributed Smart Cameras, pp. 1-6, Ghent, Belgium, Aug. 2011.
  8. P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," in Proc. 2001 IEEE Comput. Soc. Conf. Comput. Vision and Pattern Recognition, vol. 1, pp. 511-518, Hawaii, U.S.A., Dec. 2001.
  9. R. Kohavi and G. John, "Wrappers for feature subset selection," Artificial Intell., vol. 97 no. 1-2, pp. 273-324, Dec. 1997. https://doi.org/10.1016/S0004-3702(97)00043-X
  10. H. Liu, H. Motoda, and L. Yu, "Feature selection with selective sampling," in Proc. 19th Int. Conf. Mach. Learning, pp. 395-402, Sydney, Australia, July 2002.
  11. P. Viola and M. Jones, "Fast and robust classification using asymmetric AdaBoost and a detector cascade," Advances in Neural Inform. Process. Syst., vol. 14, pp. 1311-1318, Dec. 2001.
  12. P. Pudil, J. Novovicova, and J. Kittler, "Floating search methods in feature selection," Pattern Recognition Lett., vol. 15, no. 11, pp. 1119-1125, Nov. 1994. https://doi.org/10.1016/0167-8655(94)90127-9
  13. A. Treptow and A. Zell, "Combining Adaboost learning and evolutionary search to select features for real-time object detection," in Proc. IEEE Congr. Evol. Comput., pp. 2107-2113, Portland, U.S.A., June 2004.
  14. J. Kim, S. Yu, K, Toh, D. Kim, and S. Lee, "Fast on-road vehicle detection using reduced multivariate polynomial classifier," J. Korean Inst. Commun. Sci. (KICS), vol. 37A, no. 8, pp. 639-647, Aug. 2012. https://doi.org/10.7840/kics.2012.37A.8.639
  15. C. Papageorgiou, M. Oren, and T. Poggio, "A general framework for object detection," in Proc. Int. Conf. Comput. Vision, pp. 555-562, Mumbai, India, Jan. 1998.
  16. S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 2nd Ed., Prentice Hall, 2003.
  17. P. Phillips, H. Moon, P. Rauss, and S. Rizvi, "The FERET evaluation methodology for face recognition algorithms," IEEE Trans. Pattern Anal. Mach. Intell., vol. 22, no. 10, pp. 1090-1104, Oct. 2000. https://doi.org/10.1109/34.879790