基于移动对象轨迹段分割的聚类算法研究

Clustering Algorithm Based on Trajectory Segment Moving Objects Segmentation

将对象的运动轨迹作为一个整体聚类,将丢失相似子轨迹段有用的信息。为了找出相似子轨迹段,提出针对某一个轨迹进行子轨迹的聚类算法,先把长轨迹在特征点分割为一组直子段,再用基于密度的聚类算法对子段进行聚类,得到子段的可达距离排序并生成可达性图,最后根据生成的可达性图识别聚类结构。实验结果表明该方法能有效准确地识别相似子轨迹段。

The motion trajectory of the object regarded as a whole will lose a similar trajectory period of useful information.In order to find out similar orbit segments,this paper puts forward a single path of trajectory clustering algorithr.The first long trajectory in feature points is divided into a set of naoko,and a clustering algorithm based on density clustering is used to cluster,distance of the segment and the accessibility figure are generated,finally the clustering structure is identified according to the generated accessibility image.The experimental results show that the method can accurately identify similar orbit segments.