Abstract
Detecting Human objects in Thermal imagery is commonly realized in two steps. The first step is segmentation; it identifies Region of Interest (ROI) in a given image that likely to contain a human target. The second step is Classification; the identified ROI is verified for human objects. Accurate segmentation step can significantly boost classification accuracy. We present a multi-stage image segmentation framework for surveillance applications, to detect human targets present at widely varying ranges in thermal imagery. A temporal median filter is utilized to estimate the background frame from previously sampled frames. Using the estimated background image, clutters are suppressed with K-L Transformation. A box filter is applied on clutter suppressed images for identifying the Region of Interest (ROI) objects. Finally the identified ROIs are subjected to two level segmentation process for accurately extracting the silhouette.
The performance of proposed method is analysed on three thermal Infra-red video sequences containing targets in the range of 500m, 1 km and 1.5 km. This way of demonstrating performance with range is first of its kind in human target detection literature. Experimental results demonstrate that proposed segmentation approach is able to extract human silhouette without loss of significant shape information even for targets at far ranges and with background occlusions.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Arlow, H.: Thermal detection contrast of human targets. In: Proc. IEEE International Carnahan Conference on Security Technology, pp. 27–33 (1992)
Lin, S.S.: Review: Extending visible band computer vision techniques to infra-red band images. Technical Report, pp. 1-23, University of Pennsylvania, Philadelphia, Pennsylvania (2001)
Pavlidis, I., Levine, J., Baukol, P.: Thermal imaging for anxiety detection. In: Proc. IEEE Workshop on Computer Vision Beyond the Visible Spectrum: Methods and Applications, pp. 104–109 (2000)
Bhanu, B., Han, J.: Kinematic-based motion analysis in infra-red sequences. In: Proc. IEEE Workshop on Applications of Computer Vision, pp. 208–212 (2002)
Nadimi, S., Bhanu, B.: Physical models for moving shadow and object detection in video. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(8), 1079–1087 (2004)
Han, J., Bhanu, B.: Fusion of color and infrared video for moving human detection. Pattern Recognition 40(6), 1771–1784, 1079–1087 (2007)
Cielnaik, G., Duckett, T.: People Recognition by Mobile Robots. Journal of Intelligent and Fuzzy Systems 15(1), 21–27 (2004)
Fang, Y., et al.: A shape -independent method for pedestrian detection with Far-Infrared Images. IEEE Transactions on Vehicular Technology 53(6) (2004)
Conaire, C., et al.: Background modelling in Infra-red and Visible spectrum Video for people tracking. In: Proc. of IEEE International Workshop on Object Tracking & Classification Beyond the Visible Spectrum (2005)
Zhang, L., Wu, B., Nevita, R.: Pedestrian detection in infrared images based on local shape features. In: Proc. of the IEEE Computer Society Conf on Computer Vision and Pattern Recognition, pp. 1–8 (2007)
Stephen, O.H., Amber, F.: Detecting people in IR border surveillance video using scale invariant image moments. In: Proc. SPIE, vol. 7340 (2009)
Chen, B.W., Wang, W., Qin, Q.: Infrared target detection based on Fuzzy ART neural network. In: The Second Intl. Conf. on Computational Intelligence and Natural Computing, pp. 240–243. IEEE Computer Society, China (2010)
Schachter, B.J.: Target detection strategies. Opt. Eng. 52(4), 041102 (2012)
Aldriges, A., et al.: Adaptive Three – Dimensional Spatio-Temporal Filtering Techniques for Infrared Clutter Suppression. In: Proc. of SPIE on Signal and Data Processing of Small Targets, vol. 1481, pp. 110–116.
A simple Introduction to the KLT, Springer Praxis Books. Springer, Berlin (2009)
Dempster, A.P., et al.: Maximum Likelihood from incomplete data via the EM algorithm. J. Royal Stat. Soc. B 39(1), 1–38 (1977)
Wren, C.R., et al.: Real-time tracking of the human body. IEEE Transactions on Pattern Analysis and Machine Intelligence 19(7), 780–785 (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Chekuri, R.S., Prashnani, M. (2014). A Multi-stage Image Segmentation Framework for Human Detection in Mid Wave Infra-Red (MWIR) Imagery. In: Chmielewski, L.J., Kozera, R., Shin, BS., Wojciechowski, K. (eds) Computer Vision and Graphics. ICCVG 2014. Lecture Notes in Computer Science, vol 8671. Springer, Cham. https://doi.org/10.1007/978-3-319-11331-9_17
Download citation
DOI: https://doi.org/10.1007/978-3-319-11331-9_17
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-11330-2
Online ISBN: 978-3-319-11331-9
eBook Packages: Computer ScienceComputer Science (R0)