| 一种基于特征分解的图像融合方法 |
| |
| 引用本文： | 常莉红.一种基于特征分解的图像融合方法[J].浙江大学学报(理学版),2018,45(4):416. |
| |
| 作者姓名： | 常莉红 |
| |
| 作者单位： | 宁夏师范学院 数学与信息科学学院, 宁夏 固原 756000 |
| |
| 基金项目： | 国家自然科学基金资助项目（61101208）；陕西省教育厅专项（16JK1603）；重庆文理学院重点实验室开放课题项目（KFJJ1506）. |
| |
| 摘 要： | 一幅图像可以分解成几何特征不同的纹理部分和卡通部分，基于这两大特征提出了一种图像融合方法.利用卡通和纹理特征的差异，通过学习分别得到卡通字典和纹理字典.在融合过程中，分别利用特定的卡通和纹理字典对源图像的卡通和纹理部分进行融合，融合后的卡通和纹理部分经简单相加得到融合图像.实验结果表明，所提方法是有效的.
|
| 关 键 词： | 图像融合 特征分解 稀疏表示 |
| 收稿时间： | 2017-06-30 |
An image fusion method based on feature decomposition. |
| |
| CHANG Lihong.An image fusion method based on feature decomposition.[J].Journal of Zhejiang University(Sciences Edition),2018,45(4):416. |
| |
| Authors: | CHANG Lihong |
| |
| Institution: | School of Mathematics and Computer Science, Ningxia Normal University, Guyuan 756000, Ningxia Province, China |
| |
| Abstract: | An image can be decomposed into texture part and cartoon part adhering to different geometric features. An image fusion method based on these two features is proposed in this paper. A cartoon dictionary and a texture dictionary are learned from the sample images based on their different cartoon and texture features. In the fusion process, the cartoon and texture parts of the source image are fused using the specific cartoon and texture dictionary respectively. Experimental results show that the proposed method is very effective. |
| |
| Keywords: | image fusion feature decomposition sparse representation |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《浙江大学学报(理学版)》浏览原始摘要信息 |
| 点击此处可从《浙江大学学报(理学版)》下载免费的PDF全文 |
