A SIFT-based Mean Shift Algorithm for Moving Vehicle Tracking
Liang Wei,Xia Xudong,Wang Jianhua,Zhang Yi,Hu Jianming
DOI: https://doi.org/10.1109/ivs.2014.6856439
2014-01-01
Abstract:The classical mean shift algorithm is easy to pass into local maxima, which is caused by the lack of appropriate target model updating mechanism. In this paper, a SIFT-based mean shift algorithm is proposed, which can be used for continuous vehicle tracking in complex situations, such as the shape and the illumination of the vehicle object change. In our algorithm, the mean shift algorithm is utilized to determine the candidate target region, and then a judgment on the tracking effect is made according to the Bhattacharyya coefficient. If tracking fails, the candidate area is matched with the target model by SIFT feature, and a new track position is determined. Otherwise, the target model is periodically updated by SIFT feature matching, and the target model can be constantly updated according to the state change of the moving vehicle. In the scenes of moving vehicle target deformations, such as the variation of scale and illumination, the algorithm is tested and compared with other algorithms. The experimental results show that the proposed method can effectively track an object under the condition of varying illumination and shape deformation.