Directional EMD and its application to texture segmentation
文献类型：期刊论文
| 作者 | Liu, ZX ; Peng, SL
|
| 刊名 | SCIENCE IN CHINA SERIES F-INFORMATION SCIENCES
 |
| 出版日期 | 2005-06-01 |
| 卷号 | 48期号:48页码:354-365 |
| 关键词 | empirical mode decomposition (EMD)directional EMDintrinsic mode function (IMF)multi-scaletexture segmentation |
| 通讯作者 | Liu, ZX |
| 英文摘要 | In this paper we present the definition and framework of Directional Empirical Mode Decomposition (DEMD) and use DEMD to do texture segmentation. As a new technique of time-frequency analysis, EMD decomposes signals by sifting and then analyzes the instantaneous frequency of the obtained components called Intrinsic Mode Functions (IMFs). Compared with Bidimensional EMD (BEMD) which only extracts textures by radial basis function interpolation, the virtues of DEMD include: the directional quality is considered in this framework; four features can be extracted for each point from the decomposition. The technique of selecting directions for DEMD based on texture’s Wold theory is also presented. Experimental results indicate the effectiveness of the method for texture segmentation. In addition, we show the explanation for the DEMD’s ability for texture classification from visual views. |
| WOS标题词 | Science & Technology ; Technology |
| 类目[WOS] | Computer Science, Information Systems |
| 研究领域[WOS] | Computer Science |
| 关键词[WOS] | EMPIRICAL MODE DECOMPOSITION ; MULTIRESOLUTION SIGNAL DECOMPOSITION ; REPRESENTATION |
| 收录类别 | SCI |
| 语种 | 英语 |
| WOS记录号 | WOS:000230685800006 |
| 源URL | [http://ir.ia.ac.cn/handle/173211/9084]  |
| 专题 | 自动化研究所_09年以前成果 |
| 推荐引用方式 GB/T 7714 | Liu, ZX,Peng, SL. Directional EMD and its application to texture segmentation[J]. SCIENCE IN CHINA SERIES F-INFORMATION SCIENCES,2005,48(48):354-365. |
| APA | Liu, ZX,&Peng, SL.(2005).Directional EMD and its application to texture segmentation.SCIENCE IN CHINA SERIES F-INFORMATION SCIENCES,48(48),354-365. |
| MLA | Liu, ZX,et al."Directional EMD and its application to texture segmentation".SCIENCE IN CHINA SERIES F-INFORMATION SCIENCES 48.48(2005):354-365. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。