| 基于FFT与互信息准则的分层遥感图像配准 |
| |
| 引用本文： | 强赞霞,彭嘉雄. 基于FFT与互信息准则的分层遥感图像配准[J]. 计算机工程与应用, 2006, 42(36): 32-34,54 |
| |
| 作者姓名： | 强赞霞 彭嘉雄 |
| |
| 作者单位： | 1. 中原工学院,计算机科学系,郑州,450007;华中科技大学,图像识别与人工智能研究所,武汉,430074
2. 华中科技大学,图像识别与人工智能研究所,武汉,430074 |
| |
| 摘 要： | 提出了一种基于傅立叶变换和互信息准则的分层遥感图像自动配准方法。通过小波变换构造图像金字塔;在小波分解的最顶层,采用基于傅立叶变换的方法确定图像间的变换参数,并作为下一层搜索的粗略位置;然后根据互信息最大的原则逐层细化,直到最底层。试验结果表明:采用所提的方法,克服了基于互信息准则的分层配准方法耗时长的缺点,且利用分层细化的搜索策略增加了基于傅立叶变换的误差修正过程,提高了配准精度。
|
| 关 键 词： | 遥感图像 图像配准 傅立叶变换 互信息 小波变换 |
| 文章编号： | 1002-8331（2006）36-0032-03 |
| 收稿时间： | 2006-08-01 |
| 修稿时间： | 2006-08-01 |
Pyramid Approach for Remote Sensing Image Registration Based on FFT and Mutual Information |
| |
| QIANG Zan-xia,PENG Jia-xiong. Pyramid Approach for Remote Sensing Image Registration Based on FFT and Mutual Information[J]. Computer Engineering and Applications, 2006, 42(36): 32-34,54 |
| |
| Authors: | QIANG Zan-xia PENG Jia-xiong |
| |
| Abstract: | A pyramid approach for remote sensing image registration based on mutual information and FFT is presented.The image pyramid is obtained by using the wavelet transform.FFT-based technique is applied at the coarsest level of the image pyramid.Coarse-to-fine multi-resolution search approaches based on mutual information have been proposed to increase accuracy and effficiency.This algorithm has been applied on remote sensing images.The results show that this algorithm can save the computarional time and improve registration accuracy. |
| |
| Keywords: | remote sensing image image registration fourier transform mutual information wavelet transform |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
