中文摘要：

主要研究了小型无人机对地跟踪问题。针对小型无人机的盘旋状态，为了提高该情况下的目标形变，光影变化及遮挡等干扰的鲁棒性与实时性，采用基于抑制背景的MeanShift目标跟踪算法，即对目标周围和背景差异显著性较大的区域赋予较大的权值，从而把背景对目标的影响有效抑制，实现目标始终在跟踪方框内。通过架构模拟实验，该方案能够有效减小目标的连续形变以及光照变化对目标的影响；对于目标旁边景物的遮挡也有较强的抗干扰能力，而且该方案可靠性高、算法复杂度低、实时性较强。

英文摘要：

Nowadays, the target visual tracking of small unmanned aerial vehicles is intensively researched in the field of digital image processing. By analysising the process of circling flight, an effective method aiming at resolving the effects of morphological change, illumination change and occlusion is proposed, and the MeanShift tracking algorithm based on the background difference suppression adopted, thus to give greater weight to the region with significant background difference around target and make sure the goal always in the tracking box. The results obtained from indoor target tracking experiments show that the visual tracking scheme, with lower complexity and real time performance could effectively reduce the impact of the complex background on the tracked target.