论文部分内容阅读
针对机载激光雷达与航空光学影像的互补特性,提出了一种基于多源遥感数据的高精度地物信息提取和分类方法。首先从激光雷达的全波形数据获得数字高程模型(DEM)、地物的正规化数字表面模型(nDSM)和激光雷达回波相对强度信息,从航空数码相机影像获得植被指数信息;然后利用决策树方法进行地物识别。选取“黑河综合遥感联合试验”中的3种典型区域(城市、农田和水体)进行分类,结果表明:该方法能够有效地分离建筑物、高大植被、低矮植被、裸土地以及水泥地等基本地物。
Aiming at the complementary characteristics of airborne lidar and aero-optical imaging, a high-precision method of object information extraction and classification based on multi-source remote sensing data is proposed. First, the digital elevation model (DEM), the normalized digital surface model (nDSM) and the relative intensity information of lidar echoes were obtained from the full waveform data of LIDAR, and the vegetation index information was obtained from the aerial digital camera image. Then, Method for object recognition. The results showed that this method can effectively separate buildings, tall vegetation, low-lying vegetation, bare land and concrete from the three typical areas (city, farmland and water body) in the “Heihe Comprehensive Remote Sensing Joint Experiment” Other basic features.