The Journal of the Korea Contents Association (한국콘텐츠학회논문지)

The Korea Contents Association (한국콘텐츠학회)
권호 표지
DOI QR코드

DOI QR Code

Improving Weak Classifiers by Using Discriminant Function in Selecting Threshold Values

판별 함수를 이용한 문턱치 선정에 의한 약분류기 개선

  • Received : 2010.08.30
  • Accepted : 2010.12.02
  • Published : 2010.12.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose a quadratic discriminant analysis based approach for improving the discriminating strength of weak classifiers based on simple Haar-like features that were used in the Viola-Jones object detection framework. Viola and Jones built a strong classifier using a boosted ensemble of weak classifiers. However, their single threshold (or decision boundary) based weak classifier is sub-optimal and too weak for efficient discrimination between object class and background. A quadratic discriminant analysis based approach is presented which leads to hyper-quadric boundary between the object class and background class, thus realizing multiple thresholds based weak classifiers. Experiments carried out for car detection using 1000 positive and 3000 negative images for training, and 500 positive and 500 negative images for testing show that our method yields higher classification performance with fewer classifiers than single threshold based weak classifiers.

Viola와 Jones가 사용한 Haar-like 특징 기반 약분류기의 분별력을 개선하기 위하여, 2차 판별식에 기반한 판정 경계(decision boundary) 결정 방법을 제안한다. Viola와 Jones가 부스팅된 약분류기 앙상블을 사용해서 강분류기를 만들 때 사용한 단일 판정 경계 기반 약분류기는 특징 공간을 지나치게 단순하게 해석한 산물이어서 대부분의 경우 최적이 아니며, 객체 클래스와 배경 클래스 간을 효율적으로 분별하기에 흔히 너무 약하다. 이 논문에서 제안하는 2차 판별식 분석에 기반한 방법은 객체 클래스와 배경 클래스 사이에 다중 판정 경계를 사용하는 약분류기를 만들어준다. 1000개의 positive 샘플과 3000개의 negative 샘플을 훈련에 사용하고, 500개의 positive와 500개의 negative를 테스트에 사용한 차량 검출 실험을 통해서, 기존의 단일 문턱치 기반 약분류기 방식에 비해, 제안 기법이 더 적은 수의 분류기를 사용하면서도 더 우수한 분류 성능을 제공하는 것을 확인하였다.

Keywords

References

  1. P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," Conf. Computer Vision and Pattern Recognition (CVPR), 2001.
  2. P. Viola and M. Jones, "Robust Real-Time Face Detection," International Journal of Computer Vision, Vol.57, No.2, pp.137-154, 2004. https://doi.org/10.1023/B:VISI.0000013087.49260.fb
  3. Y. Freund and R. Schapire, "A decisiontheoretic generalization of on-line learning and an application to boosting," Journal of Computer and System Sciences, 1997(8).
  4. R. E. Schapire and Y. Singer, "Improved Boosting Algorithms Using Confidence-Rated Predictions," Machine Learning, Vol.37, pp.297-336, 1999. https://doi.org/10.1023/A:1007614523901
  5. S. Z. Li and Z. Q. Zhang, "FloatBoost Learning and Statistical Face Detection," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol.26, No.9, pp.1112-1123, 2004(9). https://doi.org/10.1109/TPAMI.2004.68
  6. C. Liu and H. Y. Shum, "Kullback-Leibler Boosting," Proc. Computer Vision and Pattern Recognition, pp.587-594, 2003.
  7. R. Lienhart and J. Maydt, "An Extended Set of Haar-Like Features for Rapid Object Detection," Proc. IEEE Int'l Conf. Image Processing, Vol.1, pp.900-903, 2002. https://doi.org/10.1109/ICIP.2002.1038171
  8. B. Wu, H. Ai, C. Huang, and S. Lao, "Fast Rotation Invariant Multi-View Face Detection Based on Real AdaBoost," Proc. IEEE Conf. Automatic Face and Gesture Recognition, pp.79-84, 2004.
  9. M. Ko¨lsch and M. Turk, "Robust Hand Detection," Proc. IEEE Conf. Automatic Face and Gesture Recognition, pp.614-619, 2004.
  10. T. Mita, T. Kaneko, B. Stenger and O. Hori, "Discriminative feature co-occurrence selection for object detection," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol.30, No.7, pp.1257-1269, 2008. https://doi.org/10.1109/TPAMI.2007.70767
  11. D. Zhang, S. Z. Li, and D. G. Perez, "Real-Time Face Detection Using Boosting in Hierarchical Feature Spaces," Proc. Int'l Conf. Pattern Recognition, pp.411-414, 2004.
  12. R. Duda, P. Hart, and D. Stork, Pattern Classification, Wiley- Interscience, 2000.