| 基于二维EMD的图像分层放大方法 |
| |
| 引用本文： | 田岩,黄应,彭复员,柳健. 基于二维EMD的图像分层放大方法[J]. 华中科技大学学报(自然科学版), 2007, 35(1): 20-22 |
| |
| 作者姓名： | 田岩 黄应 彭复员 柳健 |
| |
| 作者单位： | 1. 华中科技大学,电子与信息工程系,湖北,武汉,430074;华中科技大学图像信息处理与,智能控制教育部重点实验室,湖北,武汉,430074
2. 华中科技大学,电子与信息工程系湖北,武汉,430074 |
| |
| 基金项目： | 国家自然科学基金 , 国家测绘科技发展基金 |
| |
| 摘 要： | 提出了一种新的图像分层放大方法,在各层上体现图像的局部和全局的相关性,弥补了传统插值放大中的不足.其原理是先将图像用二维经验模态分解(empirical mode decomposition,EMD)方法分解为具有不同复杂程度的图层,在每一分解层上分别设计一种组合放大方法,其中组合系数密切相关于图层的复杂度,最终的放大结果再由EMD逆过程得到.实验结果和数据表明这种方法能够取得良好的图像放大效果.
|
| 关 键 词： | 图像插值 二维经验模态分解 组合插值 图像分层 方法 technique method image 效果 图像放大 数据表 实验 过程 结果 复杂度 相关性 组合系数 设计 分解层 图层 程度 empirical mode |
| 文章编号： | 1671-4512(2007)01-0020-03 |
| 修稿时间： | 2005-10-26 |
A layered image zooming method by 2D EMD technique |
| |
| Tian Yan,Huang Ying,Peng Fuyuan,Liu Jian. A layered image zooming method by 2D EMD technique[J]. JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE, 2007, 35(1): 20-22 |
| |
| Authors: | Tian Yan Huang Ying Peng Fuyuan Liu Jian |
| |
| Affiliation: | 1. Department of Eleetronies and Information Engineering;2. Image Information Proeessing and Intelligenee Control Lab. of Ministry of Education, Huazhong University of Science and Technology, Wuhan 430074, China |
| |
| Abstract: | A new layered image zooming method is proposed which can express the local and global correlation of the images on each layers to improve traditional methods.The main procedures of this method are as the following.An image was decomposed into several layers with different complexities by 2D EMD(empirical mode decomposition),each of which contains different details and roughness.Each layer was zoomed by using a combination method with different parameters which are proportional to roughness of that layer.The final zooming result was reconstructed from the combinational interpolated layers.Experimental and statistical results demonstrate that good magnified images are achieved by the proposed method. |
| |
| Keywords: | image zooming 2D EMD combination interpolation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
