| 基于SVM的遥感影像目标检测中的样本选取 |
| |
| 引用本文： | 陈亮,秦前清. 基于SVM的遥感影像目标检测中的样本选取[J]. 计算机工程与应用, 2006, 42(9): 212-214 |
| |
| 作者姓名： | 陈亮 秦前清 |
| |
| 作者单位： | 武汉大学测绘遥感信息工程国家重点实验室,武汉,430072;武汉大学测绘遥感信息工程国家重点实验室,武汉,430072 |
| |
| 摘 要： | 在基于支持向量机的遥感影像目标检测中,因为有限的目标样本和相对复杂的背景,造成检测结果的虚警率偏高。论文提出了自举算法用于样本的选取,试验表明该方法可使检测的虚警率成倍降低。
|
| 关 键 词： | 支持向量机 遥感 目标检测 样本选取 |
| 文章编号： | 1002-8331-（2006）09-0212-03 |
| 收稿时间： | 2005-07-01 |
| 修稿时间： | 2005-07-01 |
Samples Selection for RS Object Detection Based on SVMs |
| |
| Chen Liang,Qin Qianqing. Samples Selection for RS Object Detection Based on SVMs[J]. Computer Engineering and Applications, 2006, 42(9): 212-214 |
| |
| Authors: | Chen Liang Qin Qianqing |
| |
| Affiliation: | National Key Laboratory for Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University,Wuhan 430072 |
| |
| Abstract: | For RS object detection based on SVMs,the false alarm rate is commonly high because of limited object samples and relative complex background.A bootstrapping strategy is proposed to solving sample selection problem.The experiment result illustrates that the false alarm rate can be reduced to several times with the support of this method. |
| |
| Keywords: | Support Vector Machines remote sensing object detection samples selection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
