室内位置轨迹的聚类与可视化

室内移动对象轨迹数据分析是商铺促销、室内空间规划、广告竞价等具有重要商业价值的应用基础,在公共安全、应急方案中也是必不可少的部分,近年来越来越受到研究者的重视。为了实现室内移动对象轨迹聚类分析,本文提出了一种将DBSCAN聚类算法与可视化相结合的综合分析方法。首先利用DBSCAN算法对某商场大厦内采集的基于手机Wi Fi信息的室内轨迹数据进行聚类处理;然后对得到的聚类成果和信息进行分析,为该商场的布局规划和店铺调整等提供一定的参考信息;最后,对该商场大厦的室内轨迹数据进行热度图可视化展示,并将展示效果与聚类结果进行对比,相互验证。

Clustering and visualization of indoor position trajectory

The analysis of indoor mobile object trajectory data is an application foundation with important commercial value such as retail promotion, indoor space planning, and advertising bidding. It is also an indispensable part in public safety and emergency solutions. In recent years, it has received more and more attention from researchers. In order to realize the cluster analysis of indoor moving object trajectory, this paper proposes a comprehensive analysis method combining DBSCAN clustering algorithm with visualization.This paper uses DBSCAN algorithm to cluster the indoor trajectory data based on mobile phone wifi information collected in a shopping mall building, analyzes the obtained clustering results and information, and provides certain reference information for the layout planning and store adjustment of the mall.Finally, the indoor trajectory data of the mall building is visualized and displayed, and the display effect is compared with the clustering result to verify each other.