compared to the target size. Airborne targets of any
speed can be tracked as long as the object does not
cover a distance equal to half the track-gate size in
successive frames. A snapshot of a helicopter being
tracked by the system is given in Figure 4.
Apart from maintaining lock for different gate
sizes, the system was also verified to keep the target
locked while changing zoom during tracking. As the
algorithm uses target and background intensities, it
loses lock when the background and the target are
very close in intensity i.e. lighting conditions are not
good. Further, if there is more than one target in the
gate, the system will follow the target with more
contrast with the background.
4 CONCLUSIONS
A real-time visual tracking system using gray-scale
video is implemented for the specific application of
tracking airborne targets. The system is designed to
identify the target within a track gate, initialised by
the user, by rejecting the background extracted from
gate boundaries. The system is tested with different
airborne targets and speeds and is able to maintain
lock on the target provided the required lighting
conditions are maintained and the target does not
move more than half the gate size between two
successive frames. Further, provisions have been
provided for future advancements in the system for
adding auto-zoom function and algorithm selection
for different tracking scenarios.
Future research will be oriented to add motion
cue to aid the intensity based tracking system and to
improve the algorithm to adapt with different
surrounding conditions.
ACKNOWLEDGEMENTS
This work was supported and funded by
Instrumentation division of Pakistan Space and
Upper atmosphere Research Commission
(SUPARCO). We wish to thank our colleagues and
higher authorities for their support and cooperation
during the project.
REFERENCES
Avidan, S., (2007). Ensemble Tracking. IEEE
Transactions on Pattern Analysis and Machine
Intelligence, 29(2):261-271
Boni, A., Dolce, A., Rovetta, S., Zunino, R., (1996). A
Neural Networks Based Visual Tracking System. In
International Workshop on Neural Networks for
Identification, Control, Robotics, and Signal/Image
Processing (NICROSP '96), pages 128-135, Italy.
Cohen, I., and Medioni, G., (1998) Detecting and tracking
moving objects in video from and airborne observer.
In Proc. IEEE Image Understanding Workshop, pages
217--222.
Comaniciu, D., Ramesh, D., Meer, P., (2000). Real-Time
Tracking of Non-Rigid Objects Using Mean Shift, in
Proc. of 1
st
Conf. Comp. Vision Pattern Recognition,
2:142-149.
Corke, P., Hutchinson, S., (2000). Real-Time Vision,
Tracking and Control. In Proceedings of ICRA 2000,
pages 622-629
Everts, I., Sebe, N., Jones, G., (2007) Cooperative Object
Tracking with Multiple PTZ Cameras, International
Conference on Image Analysis and Processing,
Modena, Italy (to be published).
Gemignani, V., Provvedi, S., Demi, M., Paterni, M.,
Benassi, A., (1999). A DSP-Based Real Time Contour
Tracking System, in Proceedings of the 10th
International Conference on Image Analysis and
Processing, pages 630-635, Italy.
Hsu, L., Aquino, P.L.S., (1999). Adaptive visual tracking
with uncertain manipulator dynamics and uncalibrated
camera, In Proceedings of the 38th IEEE Conference
on Decision and Control, 1999. 2:1248-1253.
Li, H., Shen, C., (2006). An LMI Approach for Reliable
PTZ Camera Self-Calibration. IEEE International
Conference on Video and Signal Based Surveillance,
2006. AVSS '06. pages 79-84.
Memon, M.A., Angelov, P., and Ahmed, H., (2006). An
Approach to Real-time Color-based Object Tracking.
In 2006 International Symposium on Evolving Fuzzy
Systems, pages 82-87, UK.
Mihaylova, L., Brasnett, P., Canagarajah N., and Bull, D.,
(2007). Object Tracking by Particle Filtering
Techniques in Video Sequences, in Advances and
Challenges in Multisensor Data and Information
Processing, Vol. 8, NATO Security Through Science
Series: Information and Communication Security, E.
Lefebvre (Ed.), IOS Press, pages 260-268, the
Netherlands.
Morelande, M.R., Challa, S., (2005). Manoeuvring target
tracking in clutter using particle filters, IEEE
Transactions on Aerospace and Electronic Systems,
41(1):252 – 270.
Ribaric, S., Adrinek, G., and Segvic, S., (2004). Real-time
active visual tracking system, in Proceedings of the
12th IEEE Mediterranean Electrotechnical
Conference, pages 231-234, Dubrovnik, Croatia.
Schiele, B., and Sagerer, G., (2003). editorial: Computer
vision systems, International Journal of Machine
Vision and Applications, 14:3-4.
Shiao Y-S., (2001). Design and implementation of real-
time tracking system based on vision servo control.
Tamkang Journal of Science and Engineering. 4
(1):45-58.
VISAPP 2008 - International Conference on Computer Vision Theory and Applications
640