Journal of the Institute of Electronics Engineers of Korea SC (전자공학회논문지SC)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Traffic Object Tracking Based on an Adaptive Fusion Framework for Discriminative Attributes

차별적인 영상특징들에 적응 가능한 융합구조에 의한 도로상의 물체추적

  • Kim Sam-Yong (Dept. of Electronics and Electrical Engineering, Pohang University of Science and Technology) ;
  • Oh Se-Young (Dept. of Electronics and Electrical Engineering, Pohang University of Science and Technology)
  • 김삼용 (포항공과대학교 전자전기공학과) ;
  • 오세영 (포항공과대학교 전자전기공학과)
  • Published : 2006.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

Because most applications of vision-based object tracking demonstrate satisfactory operations only under very constrained environments that have simplifying assumptions or specific visual attributes, these approaches can't track target objects for the highly variable, unstructured, and dynamic environments like a traffic scene. An adaptive fusion framework is essential that takes advantage of the richness of visual information such as color, appearance shape and so on, especially at cluttered and dynamically changing scenes with partial occlusion[1]. This paper develops a particle filter based adaptive fusion framework and improves the robustness and adaptation of this framework by adding a new distinctive visual attribute, an image feature descriptor using SIFT (Scale Invariant Feature Transform)[2] and adding an automatic teaming scheme of the SIFT feature library according to viewpoint, illumination, and background change. The proposed algorithm is applied to track various traffic objects like vehicles, pedestrians, and bikes in a driver assistance system as an important component of the Intelligent Transportation System.

대부분의 영상을 이용한 물체추적은 적용환경을 단순화하거나 특정한 영상특징만을 적용할 수 있는 제한된 환경에서 잘 동작하기 때문에 이러한 물체추적방법은 지능자동차의 운전자보조시스템이 적용되는 복잡하고 동적인 교통 환경에서 원하는 물체를 추적하기는 어렵다. 이와 같은 물체간의 부분적인 교합이 존재하고 배경과 물체들이 동시에 동적으로 변하는 복잡한 환경에서는 물체의 색상, 외관, 외형 등과 같은 다양한 영상특징들을 적절하게 융합할 수 있는 구조가 요구된다. 본 논문에서는 기존의 파티클 필터를 이용한 적응형 융합구조[1]와 SIFT[2]를 이용한 영상특징 기술자를 강인한 영상특징으로 사용하고 시점 배경의 동적인 변화에 적응할 수 있도록 학습함으로써 추적의 강건성과 적응성을 향상시킨다. 제안된 알고리듬은 운전자 보조 시스템에서의 차량, 보행자, 자전거와 같은 도로상의 물체추적에 적용하였다.

Keywords

References

  1. A. Soto, 'A probabilistic approach for the adaptive integration of multiple visual cues using an agent framework,' Doctoral dissertation, Tech. Report CMU - RI - TR - 02 - 30, Oct. 2002
  2. D. G. Lowe, 'Distinctive image features from scale-invariant keypoints,' Int. J. Comput. Vision, vol. 60, no. 2, pp. 91 - 110, Nov. 2004 https://doi.org/10.1023/B:VISI.0000029664.99615.94
  3. D. Comaniciu, V. Ramesh, P. Meer, 'Kernel-Based Object Tracking,' IEEE T. Pattern Anal., vol. 25, issue 5, pp. 564 - 577, May 2003 https://doi.org/10.1109/TPAMI.2003.1195991
  4. C. Yang, R. Duraiswami, and L. Davis, 'Fast Multiple Object Tracking via a Hierarchical Particle Filter,' in Proc. 10th IEEE Int. Conf, Computer Vision, vol. 1, pp. 212 - 219, Oct. 2005 https://doi.org/10.1109/ICCV.2005.95
  5. K. Deguchi, O. Kawanaka, and T. Okatani, 'Object tracking by the mean-shift of regional color distribution combined with the particle-filter algorithm,' in Proc. 17th Int. Conf. Pattern Recognition, vol. 3, pp, 506 - 509, Aug. 2004 https://doi.org/10.1109/ICPR.2004.1334577
  6. T. Xiong and C. Debrunner, 'Stochastic car tracking with line- and color-based features,' IEEE T. Intell. Transp., vol. 5, no. 4, pp. 324 - 328, 2004 https://doi.org/10.1109/TITS.2004.838192
  7. F. Amell and L. Petersson, 'Generic Fusion of Visual Cues Applied to Real-World Object Segmentation,' in Proc. 2005 IEEE/RSJ Int. Conf. Intelligent Robots and Systems, pp, 2954 - 2959, 2005 https://doi.org/10.1109/IROS.2005.1545425
  8. J. Triesch, C.V.D. Malsburg, 'Self-Organized Integration of Adaptive Visual Cues for Face Tracking,' in Proc. 4th IEEE Int. Conf. on Automatic Face and Gesture Recognition, pp, 102 -107, 2000 https://doi.org/10.1109/AFGR.2000.840619
  9. S. H. Kim and I. S. Kwon, 'Scalable Representation and Learning for 3D Object Recognition using Shared Feature-based View Clustering,' in Proc. Asian Conf, on Computer Vision 2003, pp. 561 - 570, Jan. 2006
  10. B. H. Han, Y. Zhu, D. Comaniciu, and L. Davis, 'Kernel-based Bayesian filtering for object tracking,' in Proc. IEEE Computer Society Conf Computer Vision and Pattern Recognition 2005, vol. 1, pp. 227 - 234, June 2005 https://doi.org/10.1109/CVPR.2005.199
  11. 'Probabilistic algorithms in robotics,' AI Magazine vol. 22, no. 1, pp. 93 -109, 2000
  12. D. Fox, 'KLD-Sampling: Adaptive particle filter,' Advances in Neural Information Processing Systems, vol. 14, 2001
  13. V.K. Mansinghka, 'Towards Visual SLAM in Dynamic Environments,' MIT 16.412J Final ？Project Writeup, 2004
  14. Y. Ke, R. Sukthankar, 'PCA-SIFT: A More Distinctive Representation for Local Image Descriptor,' in Proc. Computer Vision and Pattern Recognition, vol. 1, pp, 511 - 517, 2004 https://doi.org/10.1109/CVPR.2004.1315206
  15. D. G. Lowe, 'Local Feature View Clustering for 3D Object Recognition,' in Proc. 2001 IEEE Computer Society Conf Computer Vision and Pattern Recognition, vol. 1, pp, 1-682 - 1-688, 2001 https://doi.org/10.1109/CVPR.2001.990541
  16. R. Bishop, Intelligent Vehicle Technology and Trends. Artech House, 2005, pp. 25 - 37
  17. S. Y. Kim, S. Y. Oh, et al. 'Front and rear vehicle detection and tracking in the day and night times using vision and sonar sensor fusion,' in Proc. 2005 IEEE/RSJ International Conf. Intelligent Robots and Systems, pp, 2173 - 2178, 2005 https://doi.org/10.1109/IROS.2005.1545321