水边线提取对揭示沿海地区自然资源管理状况和人工海域使用程度具有重要的指示作用.文章以津冀地区沿海城市海岸带为例,选取"高分一号"卫星数据为遥感数据源,首先对数据进行了辐射校正、几何校正、影像融合和裁剪等处理,然后采用面向对象的目标提取技术自动提取了研究区内大范围的水边线,并在充分了解不同海岸类型的遥感解译标志的基础上对提取的岸线进行了分类.结果显示研究区的岸线类型主要以砂砾质岸线、养殖/盐田岸线和港口码头岸线为主.在分类结果的精度验证过程中,以与"高分一号"卫星相近时期的亚米级遥感影像为底图,以目视解译的方式从中提取高精度的海岸线,并将目视解译的结果与从"高分一号"卫星数据中提取到的岸线进行对比.通过比较发现,两者位置、长度等基本吻合,从而证明了该海岸线提取方法对高分辨率遥感影像的有效性.该方法为基于"高分一号"卫星遥感数据进行大范围海岸的水边线提取提供了技术支撑,也为实现中国海岸线的定期自动更新奠定了基础.%Waterline extraction is significant to reveal the management of natural resources in coastal areas and the utilization degree of artificial coastal zones. In this research, the coastal zones in Tianjin and Hebei province are selected as the study areas. Based on GF-1 satellite data, the following processes are performed,such as radiometric correction, geometric correction, image fusion and subseting. Then, a wide range of waterline in study area is automatically extracted using object-oriented method. After that, the extracted coastline is classified based on the remote sensing interpretation symbols of different coastal types. The results show that coastline types in study area are mainly sandy coast, aquaculture/saltern coast and harbor dock coast. With contemporaneous submeter remote sensing data (Google Earth images), the coastline is also precisely extracted by visual interpretation. Finally, by comparison of the results obtained from GF-1satellite images and Google Earth images, the position and length of extracted coastline are verified, which proves that this method is effective for coastline extraction of high resolution remote sensing images. This method provides a technical support for the large-scale waterline extraction based on domestic independent high resolution remote sensing data, and lays a foundation for the accomplishing of the automatic updating of the coastline in China.
展开▼
