论文部分内容阅读
目前端元数目主要是图像处理者目视判读影像获得的经验值,往往与影像中实际典型地物种类数目不符合。根据端元数目取值不同时的分类结果均方根误差(RMSE)的大小,能够得到实际端元数,但该方法因多次分类导致计算量过大。本文在研究端元间光谱相关性与端元数目之间关系的基础上,提出一种基于端元独立性的端元数目自动获取方法。与基于RMSE的端元数目自动获取方法的对比实验表明,该方法不仅能够正确有效地获取实际端元数,提高端元提取自动化程度,而且能够大大减少运算时间。
At present, the number of terminal elements is mainly the experience gained by the image processor in visual interpretation of images, which often does not correspond with the actual number of typical types of objects in the image. According to the size of the root mean square error (RMSE) of the classification results when the number of end elements is different, the actual endmember number can be obtained, but this method causes the calculation amount to be too large due to the multiple classification. In this paper, based on the study of the relationship between the spectral correlations of terminal elements and the number of terminal elements, an automatic method for obtaining the number of terminal elements based on terminal independence is proposed. Compared with the method of automatically obtaining the number of end-units based on RMSE, the experimental results show that this method can not only acquire the actual end-elements correctly and effectively, but also improve the automation of end-point extraction, and can greatly reduce the computation time.