Advertisement

Unstructured Scene Object Localization Algorithm Based on Sparse Overcomplete Representation

  • Peng Lu
  • Yuhe Tang
  • Eryan Chen
  • Huige Shi
  • Shanshan Zhang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7389)

Abstract

Unstructured scene has many uncertainties and unpredictable states. It brings difficulties to the object localization, which is pixel-based processing. Themethod of analog visual information processing is an effective way to solve the problem mentioned above. Sparse overcomplete representation is an image representation model which is more in line with visual mechanism. However, the overcomplete representation not only increases the combinatorial search difficulty of sparse decomposition, but also changes the symmetry between input and code space. Furthermore, it makes the model solution and calculation method complicated. In order to solve the afore mentioned problem and effectively use this model to achieve automatic image object localization, this paper takes the unstructured scenes object localization as the background. Firstly, the overcomplete representation computational model which is based on energy model and score matching method is established. Then an automatic object localization method based on the neuronal response and dynamic threshold strategy is proposed and applied to the movement object localization. On this basis, the error analysis is done. Experimental results show that the method can achieve the movement object localization.

Keywords

Unstructured scenes sparse overcomplete representation object localization score matching 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Rodriguez, M., Ali, S., Kanade, T.: Tracking in Unstructured Crowded Scenes. In: IEEE 12th International Conference on Computer Vision, pp. 1389–1396 (2009)Google Scholar
  2. 2.
    Lu, P., Li, Y.Q.: Fault-image Detection Algorithm Based on Visual Perception. Chinese Journal of Scientific Instrument 31(9), 1997–2002 (2010)Google Scholar
  3. 3.
    Zheng, M., Bu, J.J.: Graph Regularized Sparse coding for Image Rrepresentation. IEEE Transactions on Image Processing 20(5), 1327–1336 (2011)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Geisler, W.S.: Visual Perception and the Statistical Properties of Natural Scenes. Annual Review of Psychology 59, 167–192 (2008)CrossRefGoogle Scholar
  5. 5.
    Thang, N.D., Rasheed, T.: Content-based Facial Image Retrieval Using Constrained Independent Component Analysis. Information Sciences 181(15), 3162–3174 (2011)CrossRefGoogle Scholar
  6. 6.
    Tseng, P.: Further Results on Stable Recovery of Sparse Overcomplete Representations in the Presence of Noise. IEEE Trans. Inf. Theory 55(2), 888–899 (2009)CrossRefGoogle Scholar
  7. 7.
    Cai, Z.M., Lai, J.Z.: An Overcomplete Learned Dictionary-Based Image Denoising Method. Acta Electronica Sinica 37(2), 347–350 (2009)Google Scholar
  8. 8.
    Mohimani, H., Babaie-Zadeh, M., Jutten, C.: A fast approach for overcomplete sparse decomposition based on smoothed L0-norm. IEEE Trans. Signal, Processing 57(1), 289–301 (2009)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Needell, D., Vershynin, R.: Uniform Uncertainty Principle and Signal Recovery via Regularized Orthogonal Matching Pursuit. Foundations of Computational Mathematics 9(3), 317–334 (2009)MathSciNetzbMATHCrossRefGoogle Scholar
  10. 10.
    Valgaerts, L., Bruhn, A.: Dense Versus Sparse Approaches for Estimating the Fundamental Matrix. Int. J. Comput Vision 96(2), 212–234 (2012)MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Joder, C., Essid, S., Richard, G.: A Conditional Random Field Framework for Robust and Scalable Audio-to-score Matching. IEEE Trans. Audio Speech Lang. Process. 19(8), 2385–2397 (2011)CrossRefGoogle Scholar
  12. 12.
    Tiesinga, P.H., Buia, C.I.: Spatial Attention in Area V4 is Mediated by Circuits in Primary Visual Cortex. Neural Networks 22(8), 1039–1054 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Peng Lu
    • 1
  • Yuhe Tang
    • 1
  • Eryan Chen
    • 1
  • Huige Shi
    • 1
  • Shanshan Zhang
    • 1
  1. 1.School of Electrical EngineeringZhengzhou UniversityZhengzhouChina

Personalised recommendations