摘要:
在对太湖、巢湖等大型湖泊进行业务化蓝藻水华遥感监测工作中,常以250 m空间分辨率的MODIS数据为主,但其像元多为水体和水华的混合像元,若用常规方法进行水华面积提取,势必会严重影响水华监测的精度和实际应用效果。针对上述问题,基于混合像元分解原理,通过混合像元分解得到水华组分在混合像元中的丰度(百分比),实现亚像元级的水华面积提取。该方法可直接根据图像的DN值进行水华面积提取,无需对数据进行辐射校正和大气校正等预处理。与常规水华提取法相比,该方法的水华面积提取精度提高了近30%。
关键词:
-
水华
/
-
亚像元
/
-
低分辨率
/
-
遥感
Abstract:
Operational bloom remote sensing monitoring usually uses MODIS data with 250 meter resolution. However, most of the remote sensing image pixels are mixture of water and algae bloom. Using images with 250 meters resolution to extract algae bloom area will seriously affect the accuracy of algal bloom monitoring. Aimed at solving this problem and based on the mixed pixel model, the authors used the decomposition of mixed pixels to extract bloom component abundance in the mixed pixels. Compared with the traditional methods, the approach proposed in this paper improves the extraction accuracy of algae bloom area by nearly 30 percent;in addition, this approach is capable of reaching the algae bloom area extraction at the sub-pixel level, thus improving the accuracy of remote sensing. In practical application,this approach can extract algae bloom area by using DN values of remote sensing image without the pre-processing of radiation and atmospheric correction for remote sensing image.
下载: