基于核相关滤波器的无人机目标跟踪算法

Object tracking algorithm based on kernelized correlation filter for unmanned aerial vehicles

多旋翼无人机已经广泛应用于各个领域,视觉目标跟踪算法对于无人机的应用具有十分重要的意义。无人机平台下的视觉目标跟踪对跟踪算法提出了很高的要求,要求在满足实时性的前提下能够在复杂场景下准确地检测跟踪目标。针对传统的核化相关滤波器跟踪算法缺乏尺度估计以及抗遮挡能力不强的问题,提出了一种改进的核化相关滤波器的目标跟踪算法。利用位置估计滤波器得到目标位置,然后采用基于颜色空间直方图的相似性度量算法得到尺度估计结果,构建小型样本库并采取自适应阈值的样本更新策略进行样本更新。提出的算法在公共数据集和无人机实拍数据集下的实验结果显示,该算法能够快速准确地跟踪目标,并且在显著遮挡和尺度变化等复杂情况下也具有较好的跟踪效果。

Unmanned aerial vehicle（ UAV） is widely used in various fields,the visual object tracking algorithm for UAV applications is of great significance. Under the unmanned aerial vehicle platform,the visual object tracking has high requirements on the tracking algorithm,and it is required to detect the tracking object accurately under the complex scene and the real-time condition. Aiming at the problem that the traditional kernelized correlation filter tracking algorithm lacks scale estimation and weak ability against occlusion,an improved visual object tracking algorithm based on kernelized correlation filter is proposed. The position estimation algorithm is used to obtain the target position. Then,the similarity measurement algorithm based on spatiogram histogram is used to get the scale estimation result. A small sample base is constructed and a sample updating strategy is adopted to update the samples. The results show that the algorithm is robust to significant occlusion and scale variant.