基于混合像元分解技术的植被信息抑制研究与应用

Research on Vegetation Information Suppression Based on Mixed Pixel Unmixing and Its Application

植被覆盖直接影响地质信息提取的精度,是遥感地质中的难点。利用混合像元分解方法可从像元中剥离出植被成分,进而获得相对纯粹的地质信息,因此具有抑制植被信息的作用。利用混合像元分解方法分解图像的植被像元组分,利用归一化植被指数变化情况描述植被抑制效果,并结合主成分分析法对实验区域进行地层界线划分。结果表明,该方法可有效划分地层界线。

Vegetation coverage directly affects the accuracy of geological information extraction,which is a difficulty in re⁃mote sensing geology.The mixed pixel unmixing method can be used to extract the vegetation components from the pixels,and obtain relatively pure geological information,so this method can suppress vegetation information.We adopt this method to decompose the vegetation pixel components of the image,and use normalized differential vegetation index change to describe the vegetation suppression effect,and then apply principal component analysis method to divide the stra⁃tum boundaries in the experimental area.The results show that the proposed method can effectively divide the stratum boundaries.