摘要
洪涝灾害是我国最主要的自然灾害类型之一,发生频率高、影响范围广,对社会经济发展和人民生命财产安全构成严重威胁。选取河南郑州“7·20”特大暴雨事件中受灾较为严重的河南扶沟县为研究区,基于洪涝灾害发生前后的Sentinel-1和Sen-tinel-2影像,利用支持向量机对灾前的Sentinel-2影像进行土地利用分类,基于Sentinel-1影像利用水体指数SDWI对灾中、灾后的水体范围进行提取,并结合GIS对研究区的灾情进行评估。结果表明:①基于支持向量机提取的土地利用分类图,总体精度达95.85%;②利用SDWI水体指数法提取的水体范围结果显示灾中、灾后的水体面积分别为36.468 km^(2)、18.770 km^(2),总体精度分别为97.6%和95.4%;③由灾情评估结果可得,曹里乡的受灾情况最为严重,最大水体变化面积达到12.63 km^(2)。
Flood disasters are the most important of the types of natural disasters with high frequency and wide impact,posing a serious threat to social and economic development and the safety of people’s lives and property.The experiment selected Fugou County of Zhoukou City,which was heavy hit by the"7.20"exceptionally heavy rainstorm event in Henan Province as the research area,selected the sentinel-1 and Sentinel-2 images before and after the flood disaster as the main data sources,used the SVM to classify the land use of the sentinel-2 image before the disaster and quickly extracted the water body area during and after the disaster on the Sentinel-1 image based on the SDWI dual polarization water index method,and combined with the GIS to evaluate the disaster situation in the research area.The results show that:①The overall accuracy of the land use classification map extracted based on the SVM is 95.85%.②The results of the water body range extracted using the SDWI water body index method show that the water body area during and after the disaster is 36.468 km^(2) and 18.770 km^(2),and the overall accuracy is 97.6%and 95.4%.③The extraction results show that Caoli Township was the most seriously affected with the maximum water body change area of 12.63 km^(2).
作者
姜晗兵
邓文彬
JIANG Hanbing;DENG Wenbin(Xinjiang University,Urumqi 830017)
出处
《中国防汛抗旱》
2024年第2期50-55,共6页
China Flood & Drought Management
基金
新疆维吾尔自治区自然科学基金(2022D01C55)。
