Multisensor fusion-based object detection and tracking using Active Shape Model

This paper proposes automatic target detection and tracking system using Active Shape Model (ASM). Existing model based approaches for tracking are either manually initiated or need some form of user interaction to locate the object in images. Also the low light environmental conditions for surveillance systems make the tracking further harder. Hence the proposed system makes use of multiple sensors in the form of IR and visible cameras to enable tracking in degraded and low light environments. The proposed algorithm consists of the following stages: (i) input image evaluation for obtaining the conditions under which the camera is placed, (ii) an integrated motion detector and target tracker, (iii) active shape tracker (AST) for performing tracking, (iv) update of tracking results for real time tracking of targets. In the first stage the input image is evaluated for the lighting conditions. If the lighting conditions are poor then IR sensor is integrated with the CCD sensor for tracking applications. In the second stage the motion detector and region tracker are used to provide feedback to AST for automatic initialization of tracking. Tracking is carried out in the third stage using ASM. The final stage extracts the parameters and tracking information and applies it to the next frame if the tracking is carried out in real time. The major contribution this work lies in the integration for a completed system, which covers from image processing to tracking algorithms. The approach of combining multiple algorithms succeeds in overcoming fundamental limitations of tracking and at the same time realizes real time implementation. Experimental results show that the proposed algorithm can track people under various environment in real-time. The proposed system has potential uses in the area of surveillance, shape analysis, and model-based coding.