| ALOS全色与多光谱影像融合的土地覆盖分类 |
| |
| 引用本文： | 吴瑞娟,何秀凤,杨智翔.ALOS全色与多光谱影像融合的土地覆盖分类[J].地理空间信息,2012(1):116-118,4,5. |
| |
| 作者姓名： | 吴瑞娟 何秀凤 杨智翔 |
| |
| 作者单位： | 河海大学地球科学与工程学院 |
| |
| 基金项目： | 国家自然科学基金资助项目(50579013);江苏科技支撑计划资助项目(BE2010316);江苏高校研究生创新计划资助项目(CX08B_110Z) |
| |
| 摘 要： | 利用Brovey、HighPass Filter和Gram-Schmidt 3种融合方法,对ALOS卫星全色与多光谱影像进行融合,并对融合后影像进行土地覆盖分类研究,从定性分析和比较融合后影像的分类精度2个方面综合评价了3种融合方法的效果。结果表明,3种融合方法都提高了影像的空间分辨率,Gram-Schmidt和HPF融合后影像光谱保持性好,同时3种融合方法不同程度上提高了影像的总体精度和Kappa系数,Gram-Schmidt最高,Brovey次之,HPF最弱,但对于不同地物分类精度又不尽相同,从整体分类结果来看,Gram-Schmidt最优。
|
| 关 键 词： | 遥感影像融合 ALOS 土地覆盖 分类 |
Land Cover Classification Based on ALOS Panchromatic and Multispectral Images Fusion |
| |
| Abstract: | Fusion of images with different spatial resolution can improve visualization of the images involved.This study tried to show that the fusion of the images could improve classification accuracy of the images.Three image fusion algorithms were employed in the study of data fusion and classification of ALOS panchromatic and multi-spectral images,taking the center part of Yancheng city as the study area.These were Brovey transform,High Pass Filter(HPF) and Gram-Schmidt(GS) methods,the effectiveness of each algorithm was evaluated based on qualitative evaluation and classification accuracy.The result showed that the GS transform was the best method in retaining spectral information of original image,which did not cause spectral distortion,and achieving the highest classification accuracy.Brovey-fused image had higher classification accuracy but significantly lost spectral properties of the original image.HPF-fused image retained spectral information of original image but had the lowest classification accuracy.Simultaneously,classification accuracy varies in different surface features.In short,the GS transform is the best method compared with the other two methods. |
| |
| Keywords: | remote sensing image fusion ALOS land cover classification |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
