基于多源遥感影像的水体要素自动提取试验

Automatic Extraction Experiment of Water Body based on Multi- Source Remote Sensing Images

常规生产项目中,水体要素的采集和更新基本采用手动或半自动方式,不但费时费力,且容易出现漏采集和错采集。本文基于多源遥感影像,采用EasyFeature、eCognition和SenseRemote软件分别进行水体要素提取试验,并对其提取精度和效率进行对比分析,得到水体要素提取的最优方法。试验结果表明,使用SenseRemote软件提取水体要素,较其他两种软件,提取精度较高,效率最高,尤其对于水体要素较多区域,优势更为明显;使用SenseRemote软件提取水体自动化程度高,提取结果基本能满足生产和设计要求,提取方法具有可推广性。

In the routine production project, manual or semi-automatic methods are used to collect and update water body, which is not only time-consuming and laborious, but also prone to leakage and error collection. based on the multi-source remote sensing images, this paper uses EasyFeature, eCognition and SenseRemote software to carry out the water body extraction experiments respectively, and compares and analyzes their extraction accuracy and efficiency to obtain the optimal method of water body extraction. The experimental results show that the SenseRemote software is more accurate and efficient than the other two software in extracting water body, especially in areas with more water body;Using Sense Remote software to extract water body is highly automated, the extraction results can basically meet the requirements of production and design, and the extraction method can be popularized.