多源图像配准技术分析与展望  

Analysis and prospect of multi-source image registration techniques

作　　者：倪国强[1] 刘琼[1] 

机构地区：[1]北京理工大学信息科学技术学院,北京100081

出　　处：《光电工程》2004年第9期1-6,共6页Opto-Electronic Engineering

基　　金：国家863计划项目(2002AA783050);国防重点预研项目(40405050401)

摘　　要：在给出多源图像配准的定义后,将常见的图像配准方法分为基于图像灰度的方法和基于图像特征的方法两大类。以此为分类基础,对国内外现有的图像配准技术和方法进行了分析和评述,并重点介绍了基于图像特征的配准方法。随后对图像配准技术所面临的主要难题作了分析。最后介绍了图像配准在军事、遥感、医学等领域的应用,并展望了其未来的发展。After giving the definition of multi-source image registration, the common image registration methods are divided into two types: intensity-based methods and feature-based methods. The analysis and evaluation for various techniques and methods of image registration at home and abroad are carried out on the basis of this two sorts, while the feature-based methods are emphasized. Then the difficulties of image registration technique are summarized and evaluated. Applications of image registration such as military, remote sensing and medicine are also introduced. At last, prospects of image registration are made.

关 键 词：图像配准 图像融合 特征提取 

分 类 号：TN911.73[电子电信—通信与信息系统]