Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Shadow Classification for Detecting Vehicles in a Single Frame

단일 프레임에서 차량 검출을 위한 그림자 분류 기법

  • Published : 2007.11.15
Download PDF
⟨ Previous Next ⟩

Abstract

A new robust approach to detect vehicles in a single frame of traffic scenes is presented. The method is based on the multi-level shadow classification, which has been shown to have the capability of extracting correct shadow shapes regardless of the operating conditions. The rationale of this classification is supported by the fact that shadow regions underneath vehicles usually exhibit darker gray level regardless of the vehicle brightness and illuminating conditions. Classified shadows provide string clues on the presence of vehicles. Unlike other schemes, neither background nor temporal information is utilized; thereby the performance is robust to the abrupt change of weather and the traffic congestion. By a simple evidential reasoning, the shadow evidences are combined with bright evidences to locate correct position of vehicles. Experimental results show the missing rate ranges form 0.9% to 7.2%, while the false alarm rate is below 4% for six traffic scenes sets under different operating conditions. The processing speed for more than 70 frames per second could be obtained for nominal image size, which makes the real-time implementation of measuring the traffic parameters possible.

본 논문에서는 단일 프레임의 교통 영상에서 차량을 검출하는 새로운 기법을 제안한다. 제안하는 기법은 동작 환경에 관계없이 여러 형태로 분류된 그림자를 추출한다. 차량의 색상과 조명 조건에 관계없이 차량이 도로와 접한 부분에는 어두운 그림자 형상을 가진다는 사실을 이용하여 그림자 분류를 수행한다. 추출된 그림자는 차량의 존재 유무를 판단할 강력한 능력을 가지고 있으며, 배경 영상과 다른 시간적 정보들을 이용하지 않으므로, 기상 및 교통 정체가 빠르게 변화하는 상황에서도 높은 검출 성능을 보장한다. 차량 위치에 존재하는 자은 정보와 그림자 영역과의 간단한 증거 추론 기법에 의해 차량을 검출할 수 있다. 6개의 다른 동작 환경의 실험에서 4% 이하의 오검출율을 보이고, 0.9%에서 7.2%의 미검출율을 보였다. 또한, 작은 크기의 영상에 대해 초당 70 프레임 이상의 처리가 가능하므로, 다양한 교통 정보를 실시간으로 측정하는 기법에 사용될 수 있다.

Keywords

References

  1. Dubuisson M, Jain AK and Taylor W, 'Segmentation and matching of vehicles in road images,' Transportation Research Record 1412, pp. 57-63, 1992
  2. Gloyer B, Aghajan HK, Siu K and Kailath T, 'Video-based freeway monitoring system using recursive vehicle tracking,' Proc. SPIE, Image and Video Processing, pp. 173-180, 1995
  3. Ikeda T, Ohnaka S and Mizoguchi M, 'Traffic Measurement with a Roadside Vision System,' IEEE Proceedings of ICPR '96, pp. 859-864, 1996
  4. Nakanishi T and Ishii K, 'Automatic vehicle image extraction based on spatio-temporal image analysis,' Proc. 11th Intern. Conf. Computer Vision, pp. 500-504, 1992
  5. Blake A, Curwen R and Zisserman A, 'A framework for spatiotemporal control in the tracking of visual contours,' Intern. Journal of Computer Vision, Vol.14, No.1, pp. 83-105, 1995 https://doi.org/10.1007/BF01421490
  6. Koller D, Daniilidis K and Nagel H, 'Model-based object tracking in monocular image sequences of road traffic scenes,' Intern. Journal of Computer Vision, Vol.10, No.3, pp. 257-281, 1993 https://doi.org/10.1007/BF01539538
  7. Koller D, Weber J and Malik J, 'Robust multiple car tracking with occlusion reasoning,' Proc. Third European Conference on Computer Vision, pp. 189-196, 1994
  8. Tan TSullivan G and Baker K, 'Pose determination and recognition of vehicles in traffic scenes,' Proc. Third European Conference on Computer Vision, pp. 501-506, 1994
  9. Roberts J and Charnley D, 'Attentive visual tracking,' Proc. British Machine Vision Conference, Guildford, UK, pp. 459-468, 1993
  10. Zhang K, 'Computation of vehicle trajectories using neural network,' Proc. British Machine Vision Conference, Guildford, UK, pp. 489-498, 1993
  11. Kollnig H and Nagel H, '3D pose estimation by directly matching polyhedral models to gray value gradients,' Intern. Journal of Computer Vision, Vol.23, No.3, pp. 283-302, 1997 https://doi.org/10.1023/A:1007927317325
  12. Smith SM and J.M. Brady JM, 'ASSET-2: Real-Time Motion Segmentation and Shape Tracking,' IEEE Trans. Pattern Analysis and Machine Intell., Vol.17, No.8, pp. 814-819, 1995 https://doi.org/10.1109/34.400573
  13. Schlutmeyer AP, et al., 'Wide area detection system (WADS),' JPL Final Rep. for FHWA, pp. 83-115, 1982
  14. Inigo RM, 'Application of machine vision to traffic monitoring and control,' IEEE Trans. on Vehicular Tech., Vol.38, No.3, pp. 112-122, 1989 https://doi.org/10.1109/25.45464
  15. Inigo RM, et al., 'Machine vision applied to vehicle guidance,' IEEE Trans. Pattern Analysis and Machine Intell., Vol.6, No.6, pp. 820-826, 1984
  16. Micalopoulos P and Wolf B, Benke R, 'Testing and field implementation on the Minnesota video detection system (AUTOSCOPE),' Transportation Research Record 1287, pp. 176-184, 1990
  17. Micalopoulos P, 'Vehicle detection through video image processing: the AUTOSCOPE system,' IEEE Trans. on Vehicular Tech., Vol.40, No.1, pp. 21-29, 1991 https://doi.org/10.1109/25.69968
  18. Pavlidis T, 'Algorithms for Graphics and Image Processing,' Computer Science Press, 1982
  19. Luong QT and Faugeras OD, 'The Fundamental matrix: theory, algorithms,' and stability analysis, International Journal of Computer Vision, Vol.17, No.1, pp. 43-75, 1996 https://doi.org/10.1007/BF00127818
  20. Park Y, 'Shape-resolving local thresholding for object detection,' Journal, Pattern Recognition Letters, Vol.22, No.8, pp. 883-890, 2001 https://doi.org/10.1016/S0167-8655(01)00034-4
  21. Pal Nr and Pal SK, 'A review on image segmentation techniques,' Pattern Recognition, Vol.26, No.9, pp. 1277-1294, 1993 https://doi.org/10.1016/0031-3203(93)90135-J