地理空间信息2024，Vol.22Issue(1)：44-48,5.DOI:10.3969/j.issn.1672-4623.2024.01.010
基于哨兵2号时序影像的水华提取
Algae Bloom Extraction Based on Sentinel-2 Time-series Images
摘要
Abstract
The phenomenon of algae bloom has caused negative impact on human life and social development,and the currently designed algae bloom indexes are difficult to be applied to Sentinel-2 images.Based on Sentinel-2 images,we analyzed the time-series spectral features of algae bloom and other features,and proposed a new algae bloom extraction method,which distinguished water bodies and water fields from the per-spective of time-series features,and then used the idea of multivariate stepwise linear regression to extract algae bloom.We conducted a compari-son experiment between this method and FAI in the study area.The results show that the extraction accuracy of method in this paper is better than FAI,the overall accuracy reaches more than 90%,and the Kappa coefficient reaches more than 0.8,which can prove the good extraction ac-curacy and universality of this method.关键词
哨兵2号/水华提取/水华指数Key words
Sentinel-2/algae bloom extraction/algae bloom index分类
天文与地球科学引用本文复制引用
敦力民,尹海丹,马嵩..基于哨兵2号时序影像的水华提取[J].地理空间信息,2024,22(1):44-48,5.基金项目
国家自然科学基金资助项目(41971295). （41971295）
