论文部分内容阅读
文章探讨提高卫星遥感大范围流域下垫面分类方法:通过集合贝叶斯分类器和决策树分类器的优势,充分利用TM影像覆被信息,结合影像时相动态信息以获得分类准确的土地覆盖/利用类型;再结合DEM生成的坡度信息得到下垫面类型,最终确定水文地理单元。以河南省的东湾流域为例进行验证,结果表明:贝叶斯法和决策树法分类各有优势,两者结合可以获取更准确的土地覆盖/利用分类结果,辅以时相类信息可以进一步修正类别间的混淆。
This paper discusses the method of improving the classification of the underlying surface in a large area of satellite remote sensing. By using the advantages of Bayesian classifier and decision tree classifier, this paper makes full use of the information of TM images and combines the dynamic information of images to obtain the accurate classification of land cover / Utilization type; combined with the slope information generated by the DEM to obtain the type of underlying surface and finally determine the hydrogeological unit. The case of Dongwan watershed in Henan Province is used as an example to verify the results. The results show that the Bayesian and decision tree methods have their own advantages. The combination of the two can obtain more accurate land cover / utilization classification results, Further amend confusion between categories.