| 基于曲波变换的红外/可见光图像融合 |
| |
| 引用本文： | 张彬,许廷发,倪国强.基于曲波变换的红外/可见光图像融合[J].计算机仿真,2008,25(11). |
| |
| 作者姓名： | 张彬 许廷发 倪国强 |
| |
| 作者单位： | 1. 中国传媒大学理学院,北京,100024
2. 北京理工大学光电工程系,北京,100081 |
| |
| 基金项目： | 中国传媒大学学科建设基金
|
| |
| 摘 要： | 和小波变换相比较,曲波变换能更好地表示图像的边缘信息.在此基础上给出了一种基于曲波变换的图像融合方法,并将其应用于红外和可见光图像融合.首先,对红外图像和可见光图像分别进行曲波变换,得到两幅图像的低频分量和不同尺度的高频分量.在对源图像的各分量融合时,对低频分量采用平均加权进行融合.对高频分量采用取绝对值较大的方法进行融合,得到融合后的低频分量和不同尺度的高频分量,最后对这些融合后的分量进行重构,得到融合图像.仿真结果表明:和基于小波变换的融合算法相比较,该算法较好地保留了源图像的细节信息,提高了融合的效果.
|
| 关 键 词： | 图像融合 脊波变换 曲波变换 小波变换 |
The Fusion of Infrared and Visible Image with Curvelet Transform |
| |
| ZHANG Bin,XU Ting-fa,NI Guo-qiang.The Fusion of Infrared and Visible Image with Curvelet Transform[J].Computer Simulation,2008,25(11). |
| |
| Authors: | ZHANG Bin XU Ting-fa NI Guo-qiang |
| |
| Affiliation: | ZHANG Bin1,XU Ting-fa2,NI Guo-qiang2(1.School of Science,The Communication University of China,Beijing 100024,China,2.Department of Optoelectironic Engineering,Beijing Institute of Technology,Beijing 100081,China) |
| |
| Abstract: | Curvelet transform can represent detail information of image very well compared with wavelet transform,and a novel fusion method based on curvelet transform is proposed.First the low frequency components and high frequency components at different scales from original images were extracted by multi-scale locally ridgelet transform.Then the fused low frequency components and high frequency components were obtained by taking weight mean and max absolute value respectively.The fusion image was reconstructed by ... |
| |
| Keywords: | Image fusion Ridgelet transform Curvelet transform Wavelet transform |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
