摘要
针对城市暴雨积涝难以实时直接监测的问题,该文提出基于普遍存在的视频监控数据对城市内涝进行实时监测的方法。基于图像差分方法,基本消除影像上非积涝区的干扰,实现对积涝区的有效识别;基于区域分割方法,较好地平衡光照和环境的变化对积涝识别的影响,实现积涝区域的精确提取;基于提取的积涝区域,采用叠加积水深度标志点的方法,直接有效地获取积涝区域的积水深度。采用南京信息工程大学一处积水情况进行实验,结果表明,该方法能够对城市积涝区进行实时动态监测,实现城市内涝精细化监测预警。
In view of the difficulty of monitoring urban waterlogging in real time,a method of urban waterlogging monitoring based on urban video image was proposed.The dry area on the image was eliminated and the waterlogging area was strengthened based on the image difference method,the region segmentation method was used to extract the waterlogging area accurately,which could reduce the influence of the change of illumination and environment,and then the method of overlapping depth mark points was used to obtain the water depth of the waterlogging area directly and effectively.Experiment results on a waterlogging area of Nanjing university of information science and technology campus showed that the method could be used to monitor and early warn urban waterlogging in real time.
作者
薛丰昌
宋肖依
唐步兴
朱一晗
王伟
XUE Fengchang;SONG Xiaoyi;TANG Buxing;ZHU Yihan;WANG Wei(School of Geography and Remote Sensing,Nanjing University of Information Science & Technology,Nanjing 210044,China)
出处
《测绘科学》
CSCD
北大核心
2018年第8期50-55,61,共7页
Science of Surveying and Mapping
基金
江苏省自然科学基金项目(BK20151458)
江苏省博士后科研基金项目(1101024B)
中国气象局北京城市气象研究城市气象科学研究基金项目(IUMKY&UMRF201103)
江苏省研究生培养创新工程项目(KYCX17_0890)
关键词
城市内涝
监测预警
视频影像
图像计算
urban waterlogging
monitoring
video image
image calculation