引用本文: | 雷琳,蒋咏梅,匡纲要.一种基于图像分类的遥感图像配准方法.[J].国防科技大学学报,2004,26(2):35-40.[点击复制] |
LEI Lin,JIANG Yongmei,KUANG Gangyao.A Method of the Remote Sensing Image Registration Based on Image Classification[J].Journal of National University of Defense Technology,2004,26(2):35-40[点击复制] |
|
|
|
本文已被:浏览 8374次 下载 6357次 |
一种基于图像分类的遥感图像配准方法 |
雷琳, 蒋咏梅, 匡纲要 |
(国防科技大学 电子科学与工程学院,湖南 长沙 410073)
|
摘要: |
遥感图像配准是遥感图像融合中最重要的预处理步骤。寻找适应性强、精度高、计算快的配准算法一直是研究的核心问题。在研究现有配准算法的基础上,提出了一种基于图像分类的特征匹配方法。该方法在基于控制点的多项式粗配准的基础上,利用分类图像相关实现了基于仿射变换模型的精配准。实验结果表明:对于同质和异质遥感图像,此方法的配准精度都达到了子像素级;而计算量与传统的基于灰度相关的方法和基于特征匹配的方法相比则大为减少。 |
关键词: 图像分类 粗配准 精配准 相关 |
DOI: |
投稿日期:2003-10-20 |
基金项目:国家部委预研项目(413220202) |
|
A Method of the Remote Sensing Image Registration Based on Image Classification |
LEI Lin, JIANG Yongmei, KUANG Gangyao |
(College of Electronic Science and Engineering, National Univ. of Defense Technology, Changsha 410073, China)
|
Abstract: |
The remote sensing image registration is one of the most important issues of image fusion. Researchers always focus onsearchingadaptive, accurate and fast registration algorithm. A feature matchingalgorithm based on classification is proposed, which uses the correlation of the classified images to realize a precise registration after a coarseregistration.Experimentresults show thefeasibility andadaptability of the method. To bothhomogeneous images andheterogeneous images, theregistrationresults reachsub-pixelprecision.Thecost ofcomputationis farless than that of theconventionalmethodsthroughgray-basedcorrelation orfeature-basedmatching. |
Keywords: image classification coarse registration precise registration correlation |
|
|
|
|
|