基于AI的突发人流聚集区域识别与预警方法

本文提出了一种突发人流聚集区域智能识别与预警的方法,采用神经网络算法对小区未来用户数进行预测,依靠K-MEANS聚类算法和阿波罗尼奥斯定理确定小区的基准值,并通过线性回归算法对预测值和基准值进行对比,确定预警小区并根据预警小区的经纬度,构建Delaunay三角网,确定预警小区基站的覆盖区域,结合邻区关系、经纬度信息对周边站点汇聚,并借助网络爬虫技术获取基站的覆盖边界,输出预警区域。

AI - based method for identification and early warning of sudden crowd gathering area

This paper proposes a sudden large area intelligent identification and early warning method, using the algorithm of neural network to predict the future users village, relying on the K MEANS clustering algorithm and Apollo mourinho"s basic value theorem to determine the area of and through linear regression algorithm for comparison between the predicted value and basic value, determine the early warning area according to the longitude and latitude of the warning area, build the Delaunay triangulation, determine the early warning area of the base station coverage, neighbourhood relations, the latitude and longitude information on surrounding site together, using the crawler technology for base station coverage boundary, output warning area.