基于光谱特征空间的监督分类中包络线去除效能评价

摘要

The continuum removal method was applied on the PROBA CHRIS hyperspectral remote sensing image of the coastal w etland in the Yellow River Estuary .Six classical supervised classification methods were implemented on the image before and after the continuum removal transformation for the land cover classification ,and then the classification results were compared by artificial interpretation and quantitative analysis .The aim of this research is to evaluate the effect of the continuum removal transformation on the supervised classification .Experimental results show that ,the continuum removal transformation is capable of improving the classification ability of certain supervised classification algorisms in the coastal w etlands classification by hyperspectral images .But the continuum removal method cannot solve the issue of mis‐classification between the bare beach and turbid water ,which generally co‐exist in the coastal wetlands and share the similar characteristics .Therefore it could not improve the overall classification accuracy of the supervised classification methods on CHRIS hyperspectral images .%对覆盖黄河口滨海湿地的PROBA CHRIS高光谱遥感影像进行包络线去除变换，采用6种常用的基于光谱特征空间的监督分类算法对变换前后的影像数据进行滨海湿地典型地物分类，通过目视对比分析和定量分析相结合的方法分析比较变换前后的分类结果，评价包络线去除方法对该类算法影响的效果和能力。结果表明，包络线去除方法能够提高部分监督分类算法针对滨海湿地典型植被类型的区分和识别能力；但由于滨海湿地内具有面积较大的裸滩和浑浊水体，这两类地物在影像中的光谱特征相近，而包络线去除方法并不能解决二者的误分问题，因此并不能提高该类算法针对CHRIS高光谱遥感影像的总体分类精度。