| 基于Q型因子分析的训练样本的选择 |
| |
| 引用本文： | 虞欣,郑肇葆.基于Q型因子分析的训练样本的选择[J].测绘学报,2007,36(1):67-71. |
| |
| 作者姓名： | 虞欣 郑肇葆 |
| |
| 作者单位： | 武汉大学,遥感信息工程学院,湖北,武汉,430079 |
| |
| 基金项目： | 国家自然科学基金资助项目(4027109440571102) |
| |
| 摘 要： | 提出基于Q型因子分析的训练样本的选择方法。它可自动地选择训练样本,而且被选择的训练样本具有代表性,使其在自动分类中充分利用所采集的样本信息,以便得到满意的分类结果。通过实验与分析证明,该方法是可行的,它优于基于人工随机地选择训练样本的方式,而且可获得更高的分类精度。
|
| 关 键 词： | 因子分析 训练样本 影像分类 |
| 文章编号： | 1001-1595（2007）01-0067-05 |
| 修稿时间： | 2006-04-112006-12-13 |
Selcection of Training Samples Based on Q-Factor Analysis |
| |
| YU Xin,ZHENG Zhao-bao.Selcection of Training Samples Based on Q-Factor Analysis[J].Acta Geodaetica et Cartographica Sinica,2007,36(1):67-71. |
| |
| Authors: | YU Xin ZHENG Zhao-bao |
| |
| Institution: | School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China |
| |
| Abstract: | A new method about selction of training samples based on Q-factor analysis is proposed,which can automatically select representative and integral training samples for capturing sufficient samples information and improve classification accuracy.The experimental results demonstrate its feasibility and it performs better than manual selection of training samples. |
| |
| Keywords: | factor analysis training samples image classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
