CSEE 2011: Advances in Computer Science, Environment, Ecoinformatics, and Education pp 55-61 | Cite as
Infrared Target Detection Based on Spatially Related Fuzzy ART Neural Network
Abstract
In order to solve the ghosts, the halo effect as well as the lower signal-to-noise ratio problems more effectively, this paper presents a spatially related fuzzy ART neural network. We introduce a laterally-inspirited learning mode into the background modeling stage. At first, we combine the region-based feature with the intensity-based feature to train the spatially related fuzzy ART neural network by the laterally-inspirited learning mode. Then two spatially related fuzzy ART neural networks are configured as master-slave pattern to build the background models and detect the infrared targets alternately. Experiments have been carried out and the results demonstrate that the proposed approach is robust to noise, and can eliminate the ghosts and the halo effect effectively. It can detect the targets effectively without much more post-process.
Keywords
infrared target detection laterally-inspirited learning mode spatially related fuzzy ART neural network master-slave patternPreview
Unable to display preview. Download preview PDF.
References
- 1.Davis, J.W., Sharma, V.: Background subtraction in thermal imagery using contour saliency. International Journal of Computer Vision 71, 161–181 (2007)CrossRefGoogle Scholar
- 2.Lin, S.-S.: Review: Extending visible band computer vision techniques to infrared band images. Technical Report, University of Pennsylvania (2001)Google Scholar
- 3.Li, Z., Bo, W., Ram, N.: Pedestrian detection in infrared imaged based on local shape features. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE Press, Minneapolis (2007)Google Scholar
- 4.Kai, J., Michael, A.: Feature based person detection beyond the visible spectrum. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 30–33. IEEE Press, Miami (2009)Google Scholar
- 5.Stephen, O.H., Ambe, F.: Detecting People in IR Border Surveillance Video Using Scale invariant image moments. In: Proceedings of SPIE, pp. 73400L-1–6. SPIE, Orlando (2009)Google Scholar
- 6.Fida, E.B., Thierry, B., Bertrand, V.: Fuzzy Foreground Detection for Infrared Videos. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, pp. 1–6. IEEE Press, Anchorage (2008)Google Scholar
- 7.Chen, B., Wang, W., Qin, Q.: Infrared target detection based on Fuzzy ART neutral network. In: The Second International Conference on Computational Intelligence and Natural Computing, pp. 240–243. IEEE Press, Wuhan (2010)Google Scholar
- 8.Lucia, M., Alfredo, P.: A self-organizing approach to background subtraction for visual surveillance applications. IEEE Transactions on Image Processing 17, 1168–1177 (2008)MathSciNetCrossRefGoogle Scholar
- 9.Object Tracking and Classification in and Beyond the Visible Spectrum, http://www.cse.ohio-state.edu/otcbvs-bench/
- 10.Kyunqnam, K., Chalidabhonqse, T.H., David, H., et al.: Real-time foreground-background segmentation using codebook model. Image Segmentation 11, 172–185 (2005)Google Scholar