军事知识服务平台
融合多源地理数据的房屋人口空间分布快速估算
Rapid Assessment of the Spatial Distribution of Population on Building Scale through Multi-source Geographic Data Integration
摘要 基于我国第七次全国人口普查数据、云南省第一次全国自然灾害风险普查房屋建筑基础数据以及遥感与社会感知地理大数据,探索单体住宅房屋人口空间分布的快速估算方法.通过线性回归与随机森林方法,筛选提取房屋尺度上与人口分布相关的关键特征因子,从建筑物基础属性(层数、建造年代)、人口活跃强度(夜间灯光)、周边生活服务设施配套(兴趣点)3 个方面,构建熵权-优劣解距离(TOPSOS)单体住宅房屋吸引力模型.结合房屋建筑面积建立乡镇(街道)人口权重分配模型,实现昭通市各乡镇(街道)单体住宅房屋人口的快速客观量化分配.结果表明:对于非行政驻地区域,估算精度较为理想,相对百分比误差在 20%左右;同时,对2020 年巧家5.0 级地震灾害受灾人口进行估算,相对百分比误差为10.06%,这说明该模型在较大范围的估算结果相对可靠.
Abstract:
Based on the data from the Seventh National Population Census of China,the basic building data from the First National Comprehensive Risk Survey of Natural Disasters in Yunnan Province,and the Remote Sensing and Social Sensing Geospatial Big Data,a rapid assessment method for the spatial distribution of population in residen-tial buildings is explored in this study.By linear regression and the Random Forest method,key factors correlated with population distribution on the building scale are identified and extracted from three aspects:building attributes(including number of floors,construction year),activity level of people(described by nighttime lights),and daily life service facilities(described by Point of Interest,POI).Taking the buildings in Zhaotong City as re-search samples,an Entropy Weight-TOPSIS model is constructed to evaluate the attractiveness of residential build-ings.Then,based on the building floor area,a population allocation weighting model is built for townships and ur-ban districts,enabling rapid,objective,and quantitative population allocation in residential buildings across all townships and urban districts in Zhaotong.The proposed method is applied into the assessment of the population in the non-administrative seat areas in Zhaotong,and the majority of Mean Absolute Percentage Error(MAPE)val-ues are approximately 20%.When the method is applied into the assessment of the population effected by the 2020 Qiaojia MS5.0 earthquake,the relative error is only 10.06%.
刊 名 地震研究 ISTICPKU
年,卷(期): 2026, 49(4)
分类号: P315.94
机标分类号: P208 TP311.13 TU984
在线出版日期: 2026年7月15日
基金项目: 2024K04:云南省地震局青年基金项目XH24039C:中国地震局地震星火计划项目CXTD202504:云南省地震局创新团队,