Object tracking based on unmanned aerial vehicle video

Abstract

The video sequence images about the ground targets of dynamic information are obtained from the high speed unmanned aerial vehicle, the camera movement has brought some difficulties to the target detection, object tracking. With the characteristics of displacement of the background in UAV video sequences images, the moving target detection and tracking technology is divided into three parts: motion estimation, target detection, object tracking. Because of the target smaller, lower resolution in the UAV video sequences image, so a region-based tracking algorithm is proposed. The computation of method, the speed of computational meets the needs of real-time tracking, experiment and achieved good results. © 2011 Springer-Verlag.