An Image-Based Tracking Algorithm for Hybrid Wireless Sensor Networks Using Epipolar Geometry

Target tracking based restrictively on acoustic and electromagnetic (EM) sensors may not provide adequate information regarding the mobile target. Hence, imaging sensors can be used to provide visual information. This paper develops an image-based tracking approach based on epipolar geometry and Kalman filtering. A corner detection technique is used to identify the prominent features in the frame sequence. By applying epipolar transforms, the trajectory of the target can be reconstructed along the image sequence. A recursive Kalman filter is then developed to enhance the robustness of the approach with regard to noise. Experimental results show that the proposed approach outperforms the existing image-based tracking approaches, especially for multi-target tracking.