Abstract

We present a robust approach to multiple-vehicle tracking, which combines deterministic and probabilistic methods. The observation model is built with an improved color correlogram, which is a feature vector with a compact correlogram using overlapping fragmentation to make the ideal form to measure similarity with a Bhattacharyya coefficient. The observation and state model of multiple vehicles is given under a CamShift framework. To overcome the disadvantage of particle impoverishment, a layered particle filter architecture embedding Camshift is proposed, which considers both the concentration and the diversity of the particles, and the particle set can better represent the posterior probability density. We also present experiments using a real video sequence to verify the proposed method.

© 2010 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
    [CrossRef]
  2. P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).
  3. K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
    [CrossRef]
  4. Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
    [CrossRef]
  5. S. T. Birchfield and S. Rangarajan, “Spatiograms versus histograms for region-based tracking,” in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (IEEE, 2005), Vol. 2, pp. 1158–1163.
  6. S. T. Birchfield and S. Rangarajan, “Spatial histograms for region-based tracking,” ETRI J. 29, 697–699 (2007).
    [CrossRef]
  7. P. Kumar, A. Dick, and M. J. Brooks, “Multiple target tracking with an efficient compact color correlogram,” in 10th International Conference on Control, Automation, Robotics and Vision (IEEE, 2008), pp. 699–704.
    [CrossRef]
  8. J. Garcia, J. Campos, and C. Ferreira, “Circular-harmonic minimum average correlation energy filter for color pattern recognition,” Appl. Opt. 33, 2180–2187 (1994).
    [CrossRef] [PubMed]
  9. O. Gualdrón, J. Nicolás, J. Campos, and M. Yzuel, “Rotation invariant color pattern recognition by use of a three-dimensional Fourier transform,” Appl. Opt. 42, 1434–1440(2003).
    [CrossRef] [PubMed]
  10. P. García-Martínez, J. Otón, J. Vallés, and H. Arsenault, “Nonlinear pattern recognition correlators based on color-encoding single-channel systems,” Appl. Opt. 43, 425–432 (2004).
    [CrossRef] [PubMed]
  11. D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2000), pp. 142–149.
    [CrossRef]
  12. N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
    [CrossRef]
  13. Q. Zhao and H. Tao, “Object tracking using color correlogram,” in 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (IEEE, 2005), pp. 263–270.
    [CrossRef] [PubMed]
  14. Q. Zhao and H. Tao, “Motion observability analysis of the simplified color correlogram for visual tracking,” in Proceedings of the 8th Asian conference on Computer Vision (Springer-Verlag, 2007), Vol. I, pp. 345–354.
  15. H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
    [CrossRef]
  16. H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.
  17. T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.
  18. G. R. Bradski, “Computer vision face tracking as a component of a perceptual user interface,” in Proceedings of the IEEE Workshop on Applications of Computer Vision (IEEE, 1998), pp. 214–219.
  19. S. Hu, G. Liang, and Z. Jing, “Robust object tracking algorithm in natural environments,” in Proceedings of the 2nd International Conference on Natural Computation (Springer, 2006), pp. 516–525.
  20. B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
    [CrossRef]

2008

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
[CrossRef]

2007

S. T. Birchfield and S. Rangarajan, “Spatial histograms for region-based tracking,” ETRI J. 29, 697–699 (2007).
[CrossRef]

2005

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

2004

2003

D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
[CrossRef]

K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
[CrossRef]

O. Gualdrón, J. Nicolás, J. Campos, and M. Yzuel, “Rotation invariant color pattern recognition by use of a three-dimensional Fourier transform,” Appl. Opt. 42, 1434–1440(2003).
[CrossRef] [PubMed]

2002

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

1994

Aittola, M.

T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.

Arsenault, H.

Birchfield, S. T.

S. T. Birchfield and S. Rangarajan, “Spatial histograms for region-based tracking,” ETRI J. 29, 697–699 (2007).
[CrossRef]

S. T. Birchfield and S. Rangarajan, “Spatiograms versus histograms for region-based tracking,” in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (IEEE, 2005), Vol. 2, pp. 1158–1163.

Bradski, G. R.

G. R. Bradski, “Computer vision face tracking as a component of a perceptual user interface,” in Proceedings of the IEEE Workshop on Applications of Computer Vision (IEEE, 1998), pp. 214–219.

Brooks, M. J.

P. Kumar, A. Dick, and M. J. Brooks, “Multiple target tracking with an efficient compact color correlogram,” in 10th International Conference on Control, Automation, Robotics and Vision (IEEE, 2008), pp. 699–704.
[CrossRef]

Campos, J.

Comaniciu, D.

D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
[CrossRef]

D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2000), pp. 142–149.
[CrossRef]

Dick, A.

P. Kumar, A. Dick, and M. J. Brooks, “Multiple target tracking with an efficient compact color correlogram,” in 10th International Conference on Control, Automation, Robotics and Vision (IEEE, 2008), pp. 699–704.
[CrossRef]

Ferreira, C.

Gangnet, M.

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

Garcia, J.

García-Martínez, P.

Gualdrón, O.

Hu, S.

S. Hu, G. Liang, and Z. Jing, “Robust object tracking algorithm in natural environments,” in Proceedings of the 2nd International Conference on Natural Computation (Springer, 2006), pp. 516–525.

Hue, C.

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

Jin, Z. H.

B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
[CrossRef]

Jing, Z.

S. Hu, G. Liang, and Z. Jing, “Robust object tracking algorithm in natural environments,” in Proceedings of the 2nd International Conference on Natural Computation (Springer, 2006), pp. 516–525.

Koller-Meier, E.

K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
[CrossRef]

Kumar, P.

P. Kumar, A. Dick, and M. J. Brooks, “Multiple target tracking with an efficient compact color correlogram,” in 10th International Conference on Control, Automation, Robotics and Vision (IEEE, 2008), pp. 699–704.
[CrossRef]

Li, C.

H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.

Liang, G.

S. Hu, G. Liang, and Z. Jing, “Robust object tracking algorithm in natural environments,” in Proceedings of the 2nd International Conference on Natural Computation (Springer, 2006), pp. 516–525.

Liu, Z.

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

Matinmikko, E.

T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.

Meer, P.

D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
[CrossRef]

D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2000), pp. 142–149.
[CrossRef]

Nicolás, J.

Nummiaro, K.

K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
[CrossRef]

Ojala, T.

T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.

Otón, J.

Peng, N. S.

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

Pérez, P.

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

Ramesh, V.

D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
[CrossRef]

D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2000), pp. 142–149.
[CrossRef]

Rangarajan, S.

S. T. Birchfield and S. Rangarajan, “Spatial histograms for region-based tracking,” ETRI J. 29, 697–699 (2007).
[CrossRef]

S. T. Birchfield and S. Rangarajan, “Spatiograms versus histograms for region-based tracking,” in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (IEEE, 2005), Vol. 2, pp. 1158–1163.

Rautiainen, M.

T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.

Sheng, H.

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.

Tao, H.

Q. Zhao and H. Tao, “Object tracking using color correlogram,” in 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (IEEE, 2005), pp. 263–270.
[CrossRef] [PubMed]

Q. Zhao and H. Tao, “Motion observability analysis of the simplified color correlogram for visual tracking,” in Proceedings of the 8th Asian conference on Computer Vision (Springer-Verlag, 2007), Vol. I, pp. 345–354.

Tian, W. F.

B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
[CrossRef]

Vallés, J.

Van Gool, L.

K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
[CrossRef]

Vermaak, J.

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

Wang, Z. W.

Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
[CrossRef]

Wei, Q.

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.

Weng, J. N.

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

Xiong, Z.

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.

Xu, Y.

Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
[CrossRef]

Yang, J.

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

Yang, X. K.

Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
[CrossRef]

Yu, S. Y.

Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
[CrossRef]

Yzuel, M.

Zhang, B.

B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
[CrossRef]

Zhang, F. C.

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

Zhao, Q.

Q. Zhao and H. Tao, “Object tracking using color correlogram,” in 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (IEEE, 2005), pp. 263–270.
[CrossRef] [PubMed]

Q. Zhao and H. Tao, “Motion observability analysis of the simplified color correlogram for visual tracking,” in Proceedings of the 8th Asian conference on Computer Vision (Springer-Verlag, 2007), Vol. I, pp. 345–354.

Appl. Opt.

ETRI J.

S. T. Birchfield and S. Rangarajan, “Spatial histograms for region-based tracking,” ETRI J. 29, 697–699 (2007).
[CrossRef]

IEEE Trans. Pattern Anal. Mach. Intell.

D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. Pattern Anal. Mach. Intell. 25, 564–577 (2003).
[CrossRef]

Image Vision Comput.

K. Nummiaro, E. Koller-Meier, and L. Van Gool, “An adaptive color-based particle filter,” Image Vision Comput. 21, 99–110(2003).
[CrossRef]

Int. J. Electron. Commun.

B. Zhang, W. F. Tian, and Z. H. Jin, “Robust appearance-guided particle filter for object tracking with occlusion analysis,” Int. J. Electron. Commun. 62, 24–32 (2008).
[CrossRef]

J. Software

N. S. Peng, J. Yang, Z. Liu, and F. C. Zhang, “Automatic selection of kernel-bandwidth for mean-shift object tracking,” J. Software 16, 1542–1551 (2005) (in Chinese).
[CrossRef]

Science in China Ser. E

H. Sheng, Z. Xiong, J. N. Weng, and Q. Wei, “An approach to detecting abnormal vehicle events in complex factors over highway surveillance video,” Science in China Ser. E 51, 199–208 (2008).
[CrossRef]

Other

H. Sheng, C. Li, Q. Wei, and Z. Xiong, “Real-time detection of abnormal vehicle events with multi-feature over highway surveillance video,” in 11th International IEEE Conference on Intelligent Transportation Systems (IEEE, 2008), pp. 550–556.

T. Ojala, M. Rautiainen, E. Matinmikko, and M. Aittola, “Semantic image retrieval with HSV correlograms,” in Proceedings of the 12th Scandinavian Conference on Image Analysis (Danish Society of Pattern Recognition and Image Analysis, 2001), pp. 621–627.

G. R. Bradski, “Computer vision face tracking as a component of a perceptual user interface,” in Proceedings of the IEEE Workshop on Applications of Computer Vision (IEEE, 1998), pp. 214–219.

S. Hu, G. Liang, and Z. Jing, “Robust object tracking algorithm in natural environments,” in Proceedings of the 2nd International Conference on Natural Computation (Springer, 2006), pp. 516–525.

Q. Zhao and H. Tao, “Object tracking using color correlogram,” in 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (IEEE, 2005), pp. 263–270.
[CrossRef] [PubMed]

Q. Zhao and H. Tao, “Motion observability analysis of the simplified color correlogram for visual tracking,” in Proceedings of the 8th Asian conference on Computer Vision (Springer-Verlag, 2007), Vol. I, pp. 345–354.

P. Kumar, A. Dick, and M. J. Brooks, “Multiple target tracking with an efficient compact color correlogram,” in 10th International Conference on Control, Automation, Robotics and Vision (IEEE, 2008), pp. 699–704.
[CrossRef]

D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2000), pp. 142–149.
[CrossRef]

Z. W. Wang, X. K. Yang, Y. Xu, and S. Y. Yu, “Camshift guided particle filter for visual tracking,” in 2007 IEEE Workshop on Signal Processing Systems (IEEE, 2007), pp. 301–306.
[CrossRef]

S. T. Birchfield and S. Rangarajan, “Spatiograms versus histograms for region-based tracking,” in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (IEEE, 2005), Vol. 2, pp. 1158–1163.

P. Pérez, C. Hue, J. Vermaak, and M. Gangnet, “Color-based probabilistic tracking,” Lect. Notes Comput. Sci. 2350, 661–675 (2002).

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (16)

Fig. 1
Fig. 1

Results of background estimation: (a) straight driveway and (b) turn driveway.

Fig. 2
Fig. 2

Initial detection region and tracking region.

Fig. 3
Fig. 3

Results of detecting the ROI of the vehicle targets using FCDT.

Fig. 4
Fig. 4

Vehicle divided into four overlapping fragments.

Fig. 5
Fig. 5

Coordinate fragment of defining color correlogram.

Fig. 6
Fig. 6

Influence of quantized level of color space k.

Fig. 7
Fig. 7

Influence of distance coefficient η.

Fig. 8
Fig. 8

Vehicle tracked when moving straight forward in the evening.

Fig. 9
Fig. 9

Vehicle tracked during turning in daytime.

Fig. 10
Fig. 10

Vehicle tracked during turning at night.

Fig. 11
Fig. 11

Deviation in each frame during turning tracked by CAMSGPF and the proposed approach in daytime: (a) centroid deviation and (b) bounding-box scale deviation.

Fig. 12
Fig. 12

Deviation in each frame during turning at night tracked by CAMSGPF and the proposed approach: (a) centroid deviation and (b) bounding-box scale deviation.

Fig. 13
Fig. 13

Two similarly colored vehicles when slight occlusion occurs in daytime: (a) tracking result, (b) centroid deviation, (c) bounding-box scale deviation.

Fig. 14
Fig. 14

Vehicles tracked when partial occlusion occurs in daytime: (a) tracking result, (b) centroid deviation, (c) bounding-box scale deviation.

Fig. 15
Fig. 15

Vehicles tracked by our approach in the evening: (a) tracking result, (b) centroid deviation of left truck, (c) bounding-box scale deviation of left truck, (d) centroid deviation of right truck, (e) bounding-box scale deviation of right truck.

Fig. 16
Fig. 16

Vehicles tracked by our approach at night: (a) tracking result, (b) centroid deviation, (c) bounding-box scale deviation.

Tables (3)

Tables Icon

Table 1 Configuration of Parameters in Experiments

Tables Icon

Table 2 Performance of CAMSGPF and Approach for Single Tracking

Tables Icon

Table 3 Performance of Tracking Multiple Vehicles (Occluded Vehicles)

Equations (32)

Equations on this page are rendered with MathJax. Learn more.

Φ c i , c j d = P ( I ( p 1 ) = c i , | p 1 p 2 | = d | I ( p 2 ) = c j ) ,
A c i d = Φ c i , c i d = P ( I ( p 1 ) = c i , | p 1 p 2 | = d | I ( p 2 ) = c i ) ,
d f ( h ) = η l f , d f ( v ) = η h f ,
f = { Φ c i , c j ( d ) f | 0 c i k , c i c j k } .
H = ( 1 , 2 , 3 , 4 ) / 4.
d ( h ) = η l , d ( v ) = η h ,
= { Φ c i , c j ( d ) | 0 c i k , c i c j k } ,
H = .
S = ( x , y , l , h ) T ,
S = ( C T , W T ) T .
{ S t S t 1 = S t 1 S t 2 + ω t ω t N ( 0 , Σ ) .
p ( S t | S t 1 , S t 2 , , S 1 ) N ( S t ; 2 S t 1 S t 2 , Σ ) .
p ( S t ( m ) | S t 1 ( m ) , , S 1 ( m ) ) N ( S t ( m ) ; 2 S t 1 ( m ) S t 2 ( m ) , Σ ( m ) ) ,
ρ = k = 1 K H * ( k ) × H ( k ) ,
D ( H * , H ) = 1 ρ .
p ( O t | S t ) e λ D 2 ( H * , H ) .
( 0 , 1 ] = ( 0 , 1 N ] ( N 1 N , 1 ] .
S i = U ( ( i 1 N , i N ] ) ,
S ¯ t i = E ( S t i ) = 2 S t 1 i S t 2 i ,
ω t i = p ( O t | S t i ) .
ms ( C ¯ t i ) = Σ j = 1 M C j m ( C j ) g ( C ¯ t i C j r 2 ) Σ j = 1 M m ( C j ) g ( C ¯ t i C j r 2 ) C ¯ t i ,
C t i ms ( C ¯ t i ) + C ¯ t i .
W t i γ M 00 ( S ¯ t i ) 256 × l ¯ t i × h ¯ t i · W ¯ t i ,
M 00 ( S ¯ t i ) = j = 1 M m ( C j ) .
i = 1 N ω t i = 1 .
Σ j = Σ j 1 e j ,
S t = i = 1 N S t i ω t i .
Err t = Dist t × Area t ,
Dist t = ( x t x t ) 2 σ x 2 + ( y t y t ) 2 σ y 2 ,
Area t = 1 Area track Area Observation min ( Area track , Area Observation ) ,
AVE = t = 1 L Err t / L ,
φ ( η t ( j ) ) = { 1 2 ( η t ( j ) ) 2 if     η ( j ) t c c | η t ( j ) | 1 2 c 2 otherwise ,

Metrics