摘要
针对现阶段基于POI数据的城市功能区划分受限于各类POI数量差别过大易出现空间耦合现象而导致部分地物无法识别的问题,本文提出一种基于人口热力数据的特定地物识别方法。以公园景点类地物为例,本文讨论人口热力数据对应公园景点类地物的时空关系特点,基于研究区域的规划图数据进行特征优选,通过随机森林模型的训练与学习来进行公园景点类地物的识别,经过精度检验后准确率达到86%,具有良好的可行性。
In order to solve the problem that the urban functional area division based on POI data is limited by the difference of different POI numbers,it is easy to appear spatial coupling phenomenon,which leads to some features can not be identified.This paper proposes a specific feature recognition method based on population thermal data.In this paper,the characteristics of the spatiotemporal relationship between the thermal data of population and the spatial-temporal relationship of the landscape features in the park are discussed.Based on the planning map data of the study area,the feature optimization is carried out.Through the training and learning of the random forest model,the recognition rate of the park scenic spots is 86%.the method is feasible.
作者
王井利
鄢士程
李昊
梁学鹏
WANG Jingli;YAN Shicheng;LI Hao;LIANG Xuepeng(School of transportation engineering,Shenyang Jianzhu University,Shenyang 110000,China)
出处
《测绘与空间地理信息》
2022年第8期5-8,共4页
Geomatics & Spatial Information Technology
关键词
人口热力数据
公园景点类地物
特征优选
随机森林模型
thermal data of population
landscape features of parks and scenic spots
feature optimization
random forest model