节点文献
基于主成分融合的遥感影像分区分类方法探讨
Discussion of Sub-Region Classification Method Based on Principal Components Fusion
【摘要】 以丹江口水库库区的典型地段为实验区,在对遥感影像进行主成分融合的基础上,采用分区分类法对影像进行分类.结果证实与传统的分类方法相比,采用该分类方法后,遥感影像的分类精度有较大幅度的提高,整幅影像的分类精度提高了近12个百分点,特别是在分区效果较好的西北山地区和东南丘陵区,分类精度提高的更多,达到16个百分点左右.
【Abstract】 In this paper, a new method, named the sub-region classification method based on principal components fusion was presented to classify remote sensing image. Comparing traditional classification method, the classification accuracy of whole research region enhances 12 percent, especially, in northwest mountains sub-region and southeast puszta sub-region, up to about 16 percent.
【关键词】 主成分融合;
分区分类;
监督分类;
非监督分类;
DEM;
NDVI;
【Key words】 principal components fusion; sub-region classification; supervised classification; unsupervised classification; DEM; NDVI;
【Key words】 principal components fusion; sub-region classification; supervised classification; unsupervised classification; DEM; NDVI;
【基金】 国家自然科学基金资助项目(30570301 & 40671175)
- 【文献出处】 河南大学学报(自然科学版) ,Journal of Henan University(Natural Science) , 编辑部邮箱 ,2007年03期
- 【分类号】TP751
- 【被引频次】11
- 【下载频次】267