Improving of Mean Shift Tracking Algorithm Using Adaptive Candidate Model

Mean shift tracking is used widely for object tracking. However, one of the main problems is that the background on a current frame has the same color as the target (object) which can reduce the correctness of tracking. This paper proposes Improving of Mean Shift Tracking Algorithm Using Adaptive Candidate (MST_AC) Model. This algorithm uses the background positions on the previous frame and the current frame to compute the new candidate model. The window size is fixed. The dataset is received from Performance Evaluation of Tracking and Surveillance (PETS) 2006 Benchmark. The experimental results show that the proposed MST_AC model receives higher correctness than those of the traditional mean shift algorithm (MS) and ICA mean shift algorithm (MS_ICA).