| 一种改进的基于互信息的自动图像配准算法 |
| |
| 引用本文： | 梅跃松,杨树兴,莫波.一种改进的基于互信息的自动图像配准算法[J].激光与红外,2007,37(5):470-473. |
| |
| 作者姓名： | 梅跃松 杨树兴 莫波 |
| |
| 作者单位： | 北京理工大学机电工程学院,北京,100081 |
| |
| 基金项目： | 国防科技应用基础研究基金 |
| |
| 摘 要： | 提出了改进图像配准的基于互信息的算法,首先将参考图像和待配准图像分解到小波域并二值化,以归一化互信息作为相似性测度进行配准.配准参数搜索作用SPSA算法和鲍威尔算法相结合的优化策略.实验证明本算法可以快速、精确且不需要人工干预地实现两幅图像的配准.
|
| 关 键 词： | 图像配准 互信息 小波分解 SPSA算法 鲍威尔法 改进 归一化互信息 自动 图像配准算法 Mutual Information Based Algorithm Image Registration Automatic 人工干预 快速 验证 优化策略 结合 鲍威尔 SPSA 作用 搜索 参数 相似性测度 |
| 文章编号： | 1001-5078（2007）05-0470-04 |
| 修稿时间： | 2006-10-252006-11-27 |
An Improved Automatic Image Registration Algorithm Based on Mutual Information |
| |
| MEI Yue-song,YANG Shu-xing,MO Bo.An Improved Automatic Image Registration Algorithm Based on Mutual Information[J].Laser & Infrared,2007,37(5):470-473. |
| |
| Authors: | MEI Yue-song YANG Shu-xing MO Bo |
| |
| Institution: | School of Mechatronic Engineering,Beijing Institute of Technology, Beijing 100081,China |
| |
| Abstract: | An improved algorithm of image registration based on mutual information was proposed,the reference image and registered image were decomposed into wavelet field and binaried. The normalized mutual information was used to be a scale of similarity to two registered images. SPSA algorithm and Powell algorithm were combined to optimize the parameter searching process. Experiments show that this algorithm can register two images rapidly, accurately and automatically. |
| |
| Keywords: | image registration mutual information wavelet decomposition SPSA algorithm Powell algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《激光与红外》浏览原始摘要信息 |
| 点击此处可从《激光与红外》下载免费的PDF全文 |
