Detection and tracking of moving object in visual tracking from a low-altitude flying helicopter

In order to perform tracking a moving object, a vision-based tracking system mounted on a low-altitude flying unmanned helicopter is established. It is most important that the moving object must accurately be detected and tracked from dynamic scene. The multi-threshold segmentation algorithm with adaptive tolerance based on grey-level histogram is presented. And then the object dynamic clustering algorithm with double sub-windows is used in order to avoid the disturbance of other similar object. The motion region is segmented in the shade by several appended thresholds for different illuminance. The object centroid position can be obtained correctly through 2~3 iterations in the process of dynamic clustering. The pan and tilt head is controlled based on the warping distance of the object centroid. Its pace is adjusted according to the direction of the object moving. The system produces good tracking results at a frame rate of 15 fps. These algorithms and methods are available, and the system has a certain extent practicability.