基于面向对象方法的天津市滨海新区湿地信息提取

Wetland information extraction from Tianjin Binhai New Area based on object-oriented method

基于天津市滨海新区2021年的Sentinel-2遥感影像数据,针对传统遥感影像湿地分类的不确定性问题,选择了面向对象分层分类的方法。采用面向对象多尺度分割算法,依据地物光谱异质性特征将遥感影像分割为光谱相似的对象,再结合不同地物的光谱指数、空间几何特征、纹理特征构建层次模型,分层提取湿地信息。分类效果同随机森林分类方法相比较,结果表明:利用面向对象的分层分类方法总体分类精度达到91.75%,Kappa系数为0.91,分类结果“斑驳现象”减少,湿地边界清晰完整。

Based on the Sentinel-2 remote sensing image data of Tianjin Binhai New Area in 2021,in view of the uncertainty of wetland classification in traditional remote sensing images,we chose the object-oriented multi-scale segmentation algorithm to segment the remote sensing image into objects with similar spectrum according to the spectral heterogeneity of ground features,and then the hierarchical model was constructed by combining the spectral index,spatial geometric features and texture features of different objects.The wetland information was extracted in layers.Compared with the random forest classification method,the overall classification accuracy of the object-oriented hierarchical classification method is 91.75%and the Kappa coefficient is 0.91."Mottled phenomenon"of classification results are reduced,and the wetland boundary is clear and complete.