崔红星, 杨红
上海海洋大学 海洋生态与环境学院, 上海 201306

摘要:
使用基于面向对象的方法提取水边线，B分量作为阈值分割条件，Sentinel-2A数据作为提取水边线的影像，通过多尺度分割与光谱差异分割组合的方式。对如东沿海的淤泥质海岸、交通围堤海岸和养殖围堤海岸3种不同类型的海岸水边线进行提取。通过提取的水边线与影像叠加，并对受潮汐影响较小的水边线做精度验证。总体来说，提取的水边线较为准确。水边线的快速准确提取，对监测海岸带动态变化具有重要意义。
关键词:Sentinel-2A海岸线面向对象B分量
DOI：10.11759/hykx20180306001
分类号:P748
基金项目:黄海绿潮业务化预测预警关键技术研究与应用（201205010）

Object-oriented waterline extraction based on Sentinel-2A satellite data, a case study in Rudong
CUI Hong-xing, YANG Hong
College of Marine Ecology and Environment, Shanghai Ocean University, Shanghai 201306, China
Abstract:
The rapid and accurate extraction of the waterline has great significance for monitoring the dynamic changes of the coastal zone. In this study, we used an object-based method to extract the waterline and the B-component as the threshold segmentation condition. We used Sentinel-2A data for our waterline extraction images, and combined multi-scale segmentation and spectral difference segmentation. We extracted three types of shoreline waterlines, i.e., silty coasts on the east coast, coasts on embankments, and those hosting aquaculture activities. After extracting the waterline from the image, we verified that the accuracy of the waterlines that are less affected by tides. In general, the extracted waterlines are accurate.
Key words:Sentinel-2Acoastlineobject-orientedB-component





PDF全文下载地址:

点我下载PDF