Using Mean-Shift Tracking Algorithms for Real-Time Tracking of Moving Images on an Autonomous Vehicle Testbed Platform

This paper describes new computer vision algorithms that have been developed to track moving objects as part of a long-term study into the design of (semi-)autonomous vehicles. We present the results of a study to exploit variable kernels for tracking in video sequences. The basis of our work is the mean shift object-tracking algorithm; for a moving target, it is usual to define a rectangular target window in an initial frame, and then process the data within that window to separate the tracked object from the background by the mean shift segmentation algorithm. Rather than use the standard, Epanechnikov kernel, we have used a kernel weighted by the Chamfer distance transform to improve the accuracy of target representation and localization, minimising the distance between the two distributions in RGB color space using the Bhattacharyya coefficient. Experimental results show the improved tracking capability and versatility of the algorithm in comparison with results using the standard kernel. These algorithms are incorporated as part of a robot test-bed architecture which has been used to demonstrate their effectiveness.

Authors:

• Benjamin Gorry
• Zezhi Chen
• Kevin Hammond
• Andy Wallace
• Greg Michaelson

Keywords:

• Hume
• functional programming
• autonomous vehicle
• pioneer robot
• vision.

References:

[1] K. Hammond and G. Michaelson, "The Hume Report, Version 0.3", 2006
[2] K. Hammond and G. Michaelson, "Hume: a Domain-Specific Language
for Real-Time Embedded Systems", Proc. of Int. Conf. on Generative
Programming and Component Engineering, Erfurt, Germany, Sept. 2003,
Springer-Verlag Lecture Notes in Comp. Sci., pp. 37-56.
[3] Y. Z. Cheng, "Mean shift, model seeking, and clustering," IEEE
Transactions on Pattern Analysis and Machine Intelligence, 17(8): 790
-799, 1995.
[4] D. Comaniciu, P. Meer, "Robust analysis of feature space: Color image
segmentation," In IEEE Conf. Computer vision and Pattern Recognition,
750 - 755, 1997.
[5] D. Comaniciu and P. Meer, "Mean shift: A robust approach toward
feature space analysis," IEEE Transactions on Pattern Analysis and
Machine Intelligence, 24(5), 603-619, 2002.
[6] D. Comaniciu, V. Ramesh, P. Meer, "Kernel-based object tracking,"
IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(5),
pp564-575, 2003.
[7] Y. Keselman and E. Micheli-Tzanakou, "Extraction and characterization
of regions of interest in biomedical images," In Proceeding of IEEE
International conference on Information Technology Application in
Biomedicine (ITAB 98), 87-90, 1998.
[8] A. Bhattacharyya, "On a measure of divergence between two statistical
populations defined by their probability distributions," Bulletin of the
Calcutta Mathematics Society, 35, pp99-110, 1943.
[9] MobileRobots Inc., "Pioneer 3 Operations Manual with MobileRobots
Exclusive Advanced Control & Operations Software", MobileRobots
Inc., January 2006.
[10] C. T. Johnston, K. T. Gribbon and D. G. Bailey. "FPGA based Remote
Object Tracking for Real-time Control", 1st International Conference on
Sensing Technology, Palmerston North, New Zealand, 2005.
[11] G. Hager and J. Peterson, "FROB: A Transformational Approach to the
Design of Robot Software", Proc. of the ninth International Symposium
of Robotics Research, Utah, USA, 1999.
[12] D. Fijma and R. Udink, "A Case Sudy in Functional Real-Time
Programming", Technical Report, Dept. of Computer Science, Univ. of
Twente, The Netherlands, 1991
[13] K. Hammond, "Exploiting Purely Functional Programming to Obtain
Bounded Resource Behaviour: the Hume Approach," First Central
European Summer School, CEFP 2005, Budapest, Hungary, July 4-15,
2005, Lecture Notes in Computer Science 4164, Springer-Verlag, 2006,
pp. 100-134.