基于混合像元分解的生物丰度指数提取

With the development of remote sensing technology, using remote sensing technology to monitor environmental has been effective. In order to obtain biological abundance index accurately, the classification base on remote sensing images must be accurate compare with ground truth data. However, the spatial resolution of remote sensing data couldn't provide every pixel with single type of land cover. Method of mixed pixel decomposition can classify this mixed pixel into percent of each land cover types. In this paper, we use linear unmixing method to get abundance of each type of land cover, calculating biological abundance index and degrees. The results show that the classification accuracy of linear unmixing method is higher than that of maximum likelihood method and we extracted a pixel as a study area of the biological abundance index.