基于网络信息与遥感影像的水库自动提取方法研究

针对传统建立水库相关数据库的方法多为线下人工统计，受到诸多因素制约的缺陷，将网络信息、电子地图及遥感影像等数据综合利用起来，实现水库自动提取。利用网络爬虫从官方水利政务网站抓取相关信息，筛选出最新的水库名称及所属地等数据，再根据水库名称及所属地，调用百度、高德等地图网站提供的应用接口，获取水库的空间坐标信息，并利用水库坐标对遥感影像做缓冲区分析，提取出水库水体范围，得到 1 个含有水库名称、坐标、面矢量图的水库基础数据库。经过验证，提取结果的查准率为 0.973 5，查全率为 0.611 3， 作为二者的调和平均值达到 0.751，能够完整监测并提取出大中型水库，但对小型水库的监测提取效果一般，可解决传统水库提取方法需要先验知识的问题，提高对水库的区域性动态监测能力。

Research on Automatic Reservoir Extraction Method Based on Network Information and Remote Sensing Image

In view of the shortcomings of traditional methods of building reservoir-related databases, which are mostly offline manual statistics and restricted by many factors, the data such as network information, electronic map and remote sensing image are comprehensively utilized, realization of automatic extraction of reservoir. Using web crawlers to grab relevant information from official water conservancy government websites, the latest data such as the name and location of the reservoir are screened, and then, according to the name and location of the reservoir, the application interface provided by the map websites such as Baidu and Gaode is used to obtain the spatial coordinate information of the reservoir, and the reservoir coordinates are used to do buffer zone analysis on the remote sensing image to extract the reservoir water body scope, a basic database of reservoir is obtained, which contains the name of reservoir, the coordinate and the vector map of reservoir. After verification, the precision rate of the extracted result is 0.9735, the recall rate is 0.6113, and the harmonic average value of the two is 0.751. The large and medium reservoirs can be completely monitored and extracted, however, the monitoring and extraction effect of small reservoirs is general. This method solves the problem that prior knowledge is needed in traditional reservoir extraction methods and improve the regional dynamic monitoring capability of reservoirs.