城市街景环境与社会安全事件的空间关系研究.pdfVIP

第18卷第8期 中国安全生产科学技术 Vol.18 No.8 2022年8月 Journal of Safety Sciencc and TechncCocy Aug. 2022 dol: 10. 11731/j. issn. 1673-193x. 2022. 08. 004 城市街景环境与社会安全事件的空间关系研究 ** 蒋耀，胡啸峰，吴建松2 (1.中国人民公安大学 信息网络安全学院，北京100038 ； 2.中国矿业大学(北京 )应急管理与安全工程学院，北京100083 ) 摘 要：为研究城市街景环境与社会安全事件之间的空间关系，提出融合语义分割技术与地理加权回归方法 ，基于国内典型城市 街景环境数据和社会安全事件数据，验证方法的可行性$研究结果表明：模型校正拟合因子C在0. 744 - 0. 841之间，拟合效果 较好，街景环境要素指标与社会安全事件存在显著空间相关性；绿视指数和天空开阔指数与社会安全事件在大部分区域呈显著 负相关关系，交通信号指数、汽车指数和人群聚集指数与社会安全事件呈显著正相关关系 ，研究结果可为城市安全规划和社会安 全风险防控提供决策支持$ 关键词：街景图像;语义分割；社会安全事件；地理加权回归 中图分类号：X915.2 文献标志码:A 文章编号：1673 - 193X( 2022) -08 - 0024 - 06 Research on spatial relationship between urban strretscape environment and societal security events JINNG Yao1 , HU Xiaofeng1 , WU Jiansong2 (1. Schoo- of Information and Cyber Security, People s Public Security University of China, Beijing 100038 , China 2. Schoo- of Emergency Manaaement and Safety Engineering, China University of Mining and Technolog1( Beoiong) ,Beoiong100083 ,Chona) Abstract: In order to study the spatial relationship belteen the urban sWeetscape environment and societal securita events , an analysis method ccupled semantic segmentation technok-y with geocraphica^y weighted reeression was proposed, which was velidated by using the sWeetscape environment data and societal securita evenW data in a typical city of China. The results showed that the ccrrection fitting factor C of this model was belteen 0. 744 and 0. 841 , which indiccted a good fitting per- formancc. Furthermore, it was found that there existed a sixnificant spatial ccrrelation be