随机森林协同Sentinel-1/2的东营市不透水层信息提取  

Application of random forest and Sentinel-1/2 in the information extraction of impervious layers in Dongying City

作　　者：刘春亭 冯权泷 金鼎坚 史同广 刘建涛 朱明水 LIU Chunting;FENG Quanlong;JIN Dingjian;SHI Tongguang;LIU Jiantao;ZHU Mingshui(School of Surveying and Geo-Informatics,Shandong Jianzhu University,Jinan 250101,China;College of Land Science and Technology,China Agriculture University,Beijing 100083,China;China Aero Geophysical Survey&Remote Sensing Center for Natural Resources,Beijing 100083,China)

机构地区：[1]山东建筑大学测绘地理信息学院,济南250101 [2]中国农业大学土地科学与技术学院,北京100083 [3]中国自然资源航空物探遥感中心,北京100083

出　　处：《自然资源遥感》2021年第3期253-261,共9页Remote Sensing for Natural Resources

基　　金：国家自然科学基金项目“多策略协同的区域乡村聚落遥感识别及演化机制和生态环境影响分析研究”(编号:42171113);国家重点研发计划-政府间国际科技创新合作重点专项“中蒙牧草多源遥感监测关键技术研发”(编号:2018YFE0122700);山东建筑大学校内博士基金“多策略协同的城市典型地表要素遥感提取”(编号:XNBS1903)共同资助。

摘　　要：不透水层是表征人类活动的重要指标,及时精确的不透水层信息对区域生态环境保护有重要意义。以山东省东营市为研究区,探索了一种基于多源Sentinel-1/2影像和随机森林的不透水层提取方法。通过对比实验发现,随机森林结合地表反射率特征、纹理特征和后向散射系数能够降低暗不透水层和亮不透水层与裸土的混淆现象,可以有效改善不透水层的估算精度(总体精度达到93.37%,Kappa系数达到0.9258)。研究结果揭示了随机森林协同Sentinel-1和Sentinel-2数据在不透水层信息提取方面有着广泛的应用前景,为融合多源数据对黄河三角洲区域遥感监测提供了参考。An impervious layer is an important indicator of human activities.Timely and accurate information of impervious layers is of great significance for the protection of the ecological environment.Taking the Yellow River Delta(Dongying City)as the study area,this study explores a novel extraction method of impervious layers by combining the random forest classification with Sentinel-1/2 data.According to comparative experiments,the confusion between dark and light impervious layers and bare soil can be reduced through the combination of the random forest algorithm with surface reflectance characteristics,texture characteristics,and backscatter coefficient,thus effectively improving the estimation accuracy of impervious layers(overall accuracy:93.37%,Kappa coefficient:0.9258).The results of this study reveal that the random forest algorithm combined with Sentinel-1/2 data is a promising approach in the information extraction of impervious layers,which will provide a reference for the remote sensing monitoring of the Yellow River Delta through the integration of multi-source data.

关 键 词：东营市 不透水层 Sentinel-2 Sentinel-1 纹理 随机森林 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]