Abstract

A method based on low rank and sparse decomposition is proposed for moving object detection by the fusion of visual and infrared video. The visual and infrared image sequences are decomposed into the joint low rank background term, the uncorrelated sparse moving nonobject term, and the common sparse moving object term via a joint minimization cost of nuclear norm, F norm, and l1 norm. This method provides a flexible framework that can easily fuse information from visual and infrared video. The prior fusion strategies are not required. The complementary information on visual and infrared images can be naturally fused in the procedure of object detection. The experimental results show that the proposed algorithm is effective.

© 2013 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. X. Bai, F. Zhou, and B. Xue, Opt. Express 19, 8444 (2011).
    [CrossRef]
  2. J. Davis and V. Sharma, Comput. Vis. Image Underst. 106, 162 (2007).
    [CrossRef]
  3. J. Han and B. Bhanu, Pattern Recogn. 40, 1771 (2007).
    [CrossRef]
  4. C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.
  5. Z. Zivkovic, in Proceedings of the IEEE Conference on Pattern Recognition (IEEE, 2004), pp. 28–31.
  6. C. Cuevas, R. Mohedano, and N. García, Opt. Lett. 37, 3159 (2012).
    [CrossRef]
  7. O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
    [CrossRef]
  8. E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
    [CrossRef]
  9. X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
    [CrossRef]
  10. V. Kolmogorov and R. Zabih, IEEE Trans. Pattern Anal. Mach. Intell. 26, 147 (2004).
    [CrossRef]
  11. A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
    [CrossRef]
  12. J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
    [CrossRef]

2013 (2)

O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
[CrossRef]

X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
[CrossRef]

2012 (2)

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

C. Cuevas, R. Mohedano, and N. García, Opt. Lett. 37, 3159 (2012).
[CrossRef]

2011 (2)

X. Bai, F. Zhou, and B. Xue, Opt. Express 19, 8444 (2011).
[CrossRef]

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

2010 (1)

J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
[CrossRef]

2007 (2)

J. Davis and V. Sharma, Comput. Vis. Image Underst. 106, 162 (2007).
[CrossRef]

J. Han and B. Bhanu, Pattern Recogn. 40, 1771 (2007).
[CrossRef]

2004 (1)

V. Kolmogorov and R. Zabih, IEEE Trans. Pattern Anal. Mach. Intell. 26, 147 (2004).
[CrossRef]

Bai, X.

Bhanu, B.

J. Han and B. Bhanu, Pattern Recogn. 40, 1771 (2007).
[CrossRef]

Boykov, Y.

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

Cai, J.

J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
[CrossRef]

Candes, E.

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
[CrossRef]

Cooke, E.

C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.

Cuevas, C.

Davis, J.

J. Davis and V. Sharma, Comput. Vis. Image Underst. 106, 162 (2007).
[CrossRef]

Delong, A.

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

García, N.

Han, J.

J. Han and B. Bhanu, Pattern Recogn. 40, 1771 (2007).
[CrossRef]

Isack, H. N.

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

Kolmogorov, V.

V. Kolmogorov and R. Zabih, IEEE Trans. Pattern Anal. Mach. Intell. 26, 147 (2004).
[CrossRef]

Li, X.

O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
[CrossRef]

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

Ma, Y.

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

Mohedano, R.

Ó Conaire, C.

C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.

O’Connor, N. E.

C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.

Oreifej, O.

O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
[CrossRef]

Osokin, A.

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

Shah, M.

O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
[CrossRef]

Sharma, V.

J. Davis and V. Sharma, Comput. Vis. Image Underst. 106, 162 (2007).
[CrossRef]

Shen, Z.

J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
[CrossRef]

Smeaton, A. F.

C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.

Wright, J.

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

Xue, B.

Yang, C.

X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
[CrossRef]

Yu, W.

X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
[CrossRef]

Zabih, R.

V. Kolmogorov and R. Zabih, IEEE Trans. Pattern Anal. Mach. Intell. 26, 147 (2004).
[CrossRef]

Zhou, F.

Zhou, X.

X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
[CrossRef]

Zivkovic, Z.

Z. Zivkovic, in Proceedings of the IEEE Conference on Pattern Recognition (IEEE, 2004), pp. 28–31.

Comput. Vis. Image Underst. (1)

J. Davis and V. Sharma, Comput. Vis. Image Underst. 106, 162 (2007).
[CrossRef]

IEEE Trans. Pattern Anal. Mach. Intell. (3)

X. Zhou, C. Yang, and W. Yu, IEEE Trans. Pattern Anal. Mach. Intell. 35, 597 (2013).
[CrossRef]

V. Kolmogorov and R. Zabih, IEEE Trans. Pattern Anal. Mach. Intell. 26, 147 (2004).
[CrossRef]

O. Oreifej, X. Li, and M. Shah, IEEE Trans. Pattern Anal. Mach. Intell. 35, 450 (2013).
[CrossRef]

Int. J. Comput. Vis. (1)

A. Delong, A. Osokin, H. N. Isack, and Y. Boykov, Int. J. Comput. Vis. 96, 1 (2012).
[CrossRef]

J. Assoc. Comput. Mach. (1)

E. Candes, X. Li, Y. Ma, and J. Wright, J. Assoc. Comput. Mach. 58, 1 (2011).
[CrossRef]

Opt. Express (1)

Opt. Lett. (1)

Pattern Recogn. (1)

J. Han and B. Bhanu, Pattern Recogn. 40, 1771 (2007).
[CrossRef]

SIAM J. Optim. (1)

J. Cai, E. Candes, and Z. Shen, SIAM J. Optim. 20, 1956 (2010).
[CrossRef]

Other (2)

C. Ó Conaire, N. E. O’Connor, E. Cooke, and A. F. Smeaton, in Proceedings of the IEEE Conference on Information Fusion (IEEE, 2006), pp. 1–7.

Z. Zivkovic, in Proceedings of the IEEE Conference on Pattern Recognition (IEEE, 2004), pp. 28–31.

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 (1)

Fig. 1.
Fig. 1.

Results (first two rows are the 100th and 300th of the OTCBVS, last two rows are the 50th and 250th of the AIC). (a) Original visual images. (b) Original infrared images. (c) MoG [5] with prior fusion rule. (d) Decolor [3] with prior fusion rule. (e) Our method without prior fusion rule. (f) Ground truths.

Tables (1)

Tables Icon

Table 1. Evaluation Results for Comparison

Equations (6)

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

argminB,D(rank(B)+λD0),s.t.F=B+D,
argminBV,BR,DV,DR(rank(BVBR)+λ(DVDR)0),s.t.(FVFR)=(BVBR)+(DVDR).
argminBV,BR,I{0,1},ZV{0,1},ZR{0,1}12[(FVFR)(BVBR)](I+ZVI+ZR)cF2+γ(BVBR)*+λI1+λVZV1+λRZR1+ξG(I,ZV,ZR)s.t.IZVF2=0,IZRF2=0,ZVZRF2=0,
argminBV,BR12[(FVFR)(BVBR)](I+ZVI+ZR)cF2+γ(BVBR)*=argminB12(FB)ScF2+γB*=argminB12WBF2+γB*,
argminI{0,1},ZV{0,1},ZR{0,1}12(ΔVΔR)(I+ZVI+ZR)cF2+λI1+λVZV1+λRZR1+ξG(I,ZV,ZR)s.t.IZVF2=0,IZRF2=0,ZVZRF2=0,
argminL={lmn{1,2,3,4}}mnU(gmn=lmn)+ξG(I,ZV,ZR),

Metrics