| 结合中心投影和非下采样Contourlet的不变特征提取 |
| |
| 引用本文： | 孙涛,吴军,李康顺. 结合中心投影和非下采样Contourlet的不变特征提取[J]. 电视技术, 2013, 37(17) |
| |
| 作者姓名： | 孙涛 吴军 李康顺 |
| |
| 作者单位： | 1. 江西理工大学信息工程学院,江西赣州,341000
2. 江西理工大学信息工程学院,江西赣州341000;华南农业大学信息学院,广东广州510642 |
| |
| 基金项目： | 国家自然科学基金项目（面上项目，重点项目，重大项目） |
| |
| 摘 要： | 在基于形状的目标识别中,提取出鉴别力强且具有不变性的特征至关重要.为了克服目前局部不变特征提取方面存在的不足,提出一种结合中心投影和非下采样Contourlet变换的全局图像不变特征提取算法.算法首先利用中心投影获得图像的广义轮廓表示,其次在相位域中对广义轮廓进行离散傅里叶变换,最后对离散傅里叶系数进行NSCT变换并构造不变特征.实验结果证明了该算法的不变性.
|
| 关 键 词： | 不变特征提取 中心投影 非下采样Contourlet变换 仿射变换 |
| 收稿时间： | 2012-11-07 |
| 修稿时间： | 2012-11-22 |
Invariant Feature Extraction Using Central Projection and Non-subsampled Contourlet Transform |
| |
| suntao,wujun and LI Kang-Shun. Invariant Feature Extraction Using Central Projection and Non-subsampled Contourlet Transform[J]. Ideo Engineering, 2013, 37(17) |
| |
| Authors: | suntao wujun LI Kang-Shun |
| |
| Affiliation: | School of Information Engineering, Jiangxi University of Science & Technology,School of Information Engineering, Jiangxi University of Science & Technology,School of Information, South China Agricultural University |
| |
| Abstract: | Extracting features which are invariant and discriminate targets with similar shape is one of key problems in shape-based target recognition. In order to solve the problems of local features, a novel global invariant feature extraction algorithm using central projection and Non-subsampled Contourlet Transform (NSCT) is proposed in this paper. Firstly, the image generalized contourlet is obtained by using central projection. Secondly, the image generalized contourlet is transformed by FFT in phase domain. Finally, the invariant feature is constriction by NSCT coefficients. Experimental results demonstrate the potential of invariant feature. |
| |
| Keywords: | invariant feature extraction central projection Non-subsampled Contourlet Transform affine transform |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《电视技术》浏览原始摘要信息 |
|
点击此处可从《电视技术》下载全文 |
