| 基于子空间的3D目标识别和姿态估计方法 |
| |
| 引用本文： | 胡薇,张桂林,任仙怡. 基于子空间的3D目标识别和姿态估计方法[J]. 红外与激光工程, 2004, 33(6): 592-596 |
| |
| 作者姓名： | 胡薇 张桂林 任仙怡 |
| |
| 作者单位： | 华中科技大学,图像识别与人工智能研究所,图像信息处理与智能控制教育部重点实验室,湖北,武汉,430074;华中科技大学,图像识别与人工智能研究所,图像信息处理与智能控制教育部重点实验室,湖北,武汉,430074;华中科技大学,图像识别与人工智能研究所,图像信息处理与智能控制教育部重点实验室,湖北,武汉,430074 |
| |
| 基金项目： | 武器预研基金项目(51476010301JW0501) |
| |
| 摘 要： | 提出一种基于子空间的3D目标识别方法。该方法对3D目标进行事先的训练学习，采集目标可能出现时的图像，提取场景中目标的主要特征成分，建立所有目标样本图像和每个目标样本图像对应的两类特征子空间，分别用来确定目标类型和姿态。当输入一幅未知的待识别目标样像，识别系统将其分别向两类特征空间投影，根据它在两类特征子空间中的投影位置并参照目标特征的分布规律识别目标类型和姿态。实验证明，该方法具备对目标多种姿态图像畸变的鲁棒性，对光照变化也有很好的抑制作用，取得良好的目标识别效果。
|
| 关 键 词： | 目标识别 子空间 特征空间 |
| 文章编号： | 1007-2276(2004)06-0592-05 |
| 收稿时间： | 2003-11-12 |
| 修稿时间： | 2003-11-12 |
Eigenspace-based approach for object recognition and pose estimation |
| |
| HU Wei,ZHANG Gui-lin,REN Xian-yi ligent Control,Institute for Pattern Recognition and Artificial Intelligence,Huazhong University of Science and Technology,Wuhan ,China). Eigenspace-based approach for object recognition and pose estimation[J]. Infrared and Laser Engineering, 2004, 33(6): 592-596 |
| |
| Authors: | HU Wei ZHANG Gui-lin REN Xian-yi ligent Control Institute for Pattern Recognition Artificial Intelligence Huazhong University of Science Technology Wuhan China) |
| |
| Affiliation: | HU Wei,ZHANG Gui-lin,REN Xian-yi ligent Control,Institute for Pattern Recognition and Artificial Intelligence,Huazhong University of Science and Technology,Wuhan 430074,China) |
| |
| Abstract: | A new image recognition algorithm—eigenspace-based approach is proposed. Firstly, recognition system obtains the possible object′s images. Then, all images and image set of each object are compressed to obtain low|dimensional subspace, called the universal eigenspace and object eigenspace separately. Given an unknown input image, the recognition system projects the image into two kinds of eigenspace. The exact position of the projection in the eigenspace and the distribution formula of samples′ projections determine the object′s species and poses in the image. Experiments are conducted using several objects with complex appearance characteristics. The results show that the proposed approach could keep robustness for the variable poses of object, and could restrain illumination variety. |
| |
| Keywords: | Target recognition Subspace Eigenspace |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
