摘要
为提高城市土地利用动态监测的精度与效率,以某市核心城区为研究区,构建多源遥感数据融合体系,集成光学影像、合成孔径雷达(synthetic aperture radar,SAR)数据、激光雷达(light detection and ranging,LiDAR)点云与社会辅助数据,通过图像分割、特征建模、多模态深度学习分类与时序变化分析,实现地类精细识别与变化趋势建模。研究表明,融合Transformer模型的分类精度达89.1%,绿地转化识别误差控制在5%以内,地类变化与城市扩张路径清晰可判。多源融合方法的应用显著提高了城市复杂用地变化的监测能力与规划决策支持水平。
In order to improve the accuracy and efficiency of urban land use dynamic monitoring,a multi-source remote sensing data fusion system is established by taking the core urban area of a city as the research area.The optical image,synthetic aperture radar(SAR)data,light detection and ranging(LiDAR)point cloud and social auxiliary data are integrated,through image segmentation,feature modeling,multi-modal deep learning classification and time sequence change analysis,the refined land type identification and change trend modeling are realized.The research shows that the classification accuracy of the Transformer-fused model reaches 89.1%,the green space conversion identification error is controlled within 5%,and the land type changes and urban expansion paths are clearly identifiable.The application of multi-source fusion method significantly improves the monitoring capability of urban complex land use changes and planning decisionmaking support level.
作者
林月卿
LIN Yueqing(Shenzhen Aihua Survey Engineering Co.,Ltd.,Shenzhen 518111,Guangdong,China)
出处
《工程技术研究》
2025年第18期220-222,共3页
Engineering and Technological Research
关键词
多源遥感
土地利用变化
数据融合
城市监测
multi-source remote sensing
land use change
data fusion
urban monitoring
