Video target tracking is a critical problem in the field of computer vision. Particle filters have been proven to be very useful in\r\ntarget tracking for nonlinear and non-Gaussian estimation problems. Although most existing algorithms are able to track targets\r\nwell in controlled environments, it is often difficult to achieve automated and robust tracking of pedestrians in video sequences\r\nif there are various changes in target appearance or surrounding illumination. To surmount these difficulties, this paper presents\r\nmultitarget tracking of pedestrians in video sequences based on particle filters. In order to improve the efficiency and accuracy of\r\nthe detection, the algorithm firstly obtains target regions in training frames by combining the methods of background subtraction\r\nand Histogram of Oriented Gradient (HOG) and then establishes discriminative appearance model by generating patches and\r\nconstructing codebooks using superpixel and Local Binary Pattern (LBP) features in those target regions. During the process of\r\ntracking, the algorithm uses the similarity between candidates and codebooks as observation likelihood function and processes\r\nsevere occlusion condition to prevent drift and loss phenomenon caused by target occlusion. Experimental results demonstrate\r\nthat our algorithm improves the tracking performance in complicated real scenarios.
Loading....