面向对象的多源数据建筑物提取

建筑物是城市的重要标志之一,综合利用LiDAR数据和高分辨率遥感影像可以充分发挥不同数据源中提取建筑物的优势。本文基于面向对象分类理论,利用机载LiDAR数据和GeoEye高空间分辨率遥感影像,在多尺度分割的基础上对实验区分类并提取建筑物,进而对提取结果进行精度评价。实验表明,将LiDAR数据与高分辨率影像数据结合能够很好地提取建筑物,建筑物提取精度达89.28%。

Building Extraction Based on Multi-Source Data

Building has been one of the important signs of the city; comprehensive utilization of LiDAR data and high resolution re-mote sensing image can give full play to the advantages of different data sources.Based on the theory of object-oriented classifica-tion, the paper classified the image and extracted the buildings after multi-scale segmentation by using GeoEye high spatial resolution remote sensing image and the LiDAR data, and then evaluated the extraction results.The Experiments show that combine LIDAR data and high resolution image data can be effective to extracted buildings, and buildings extraction accuracy is 89.28%.