A. Al-Hamadi and B. Michaelis (Germany)
Tracking, Colour space, Motion estimation, Application
Non-cooperative measuring situations (e.g. shadow) cause undesired effects in detection and tracking of moving objects. These measuring situations lead to strong modifications of the intensity values and can lead to an incorrect determination of the correspondences in a video sequence and finally to a false tracking information. In this paper, we therefore propose a technique for the automatic detection and tracking of objects in video sequences that works robustly under such situations. For this technique, we apply the colour information and neuronal recognition system to the improvement of the matching quality when disturbances occur. This system structure is adaptive and robust in comparison with the intensity-based conventional methods.
Important Links:
Go Back
