Multiple object tracking using HSV color space
Snekalatha Saravanakumar, Appachi Vadivel, Christine Ahmed · 2011
Multiple object tracking is one of the most advanced researches in the field of computer vision. Various algorithms and approaches have been proposed and most of them use both foreground and background information. In this paper, we propose a multiple object strategy along with a failure recovery mechanism. The frames are modeled as target and non-target and based on which the tracking strategy works. The pixel values in the target and non-target group is transformed to HSV color space and represented as true and gray value. While tracking multiple objects, the tracking fails during occlusion, we handle this situation by using a suitable failure detection strategy. The histograms of the target objects in the current frame and the next frame is used for failure recovery. For evaluating the performance of the proposed method, we have used PETS video sequences and compare it with some of the recently proposed method. The result is encouraging.