| 基于SVMs的大幅遥感影像目标快速检测方法 |
| |
| 引用本文： | 胡淑均. 基于SVMs的大幅遥感影像目标快速检测方法[J]. 武汉工业学院学报, 2005, 24(2): 15-18 |
| |
| 作者姓名： | 胡淑均 |
| |
| 作者单位： | 武汉工业学院,电气信息工程系,湖北,武汉,430023 |
| |
| 摘 要： | 针对高分辨率遥感影像数据幅度大、内容复杂的现状，提出了基于分层分类器的检测方法。从简化分类器的结构和缩减检测区域探讨了提高系统效率的途径。实验表明，这种新的搜索策略提高了大幅遥感影像目标检测的效率。
|
| 关 键 词： | 遥感影像 快速检测方法 高分辨率遥感 影像数据 系统效率 目标检测 搜索策略 分类器 |
| 文章编号： | 1009-4881(2005)02-0015-04 |
| 修稿时间： | 2005-03-09 |
THE RAPID METHOD OF REMOTE SENSING IMAGE DETECTION BASED ON SUPPORT VECTOR MACHINES |
| |
| HU Shu-jun. THE RAPID METHOD OF REMOTE SENSING IMAGE DETECTION BASED ON SUPPORT VECTOR MACHINES[J]. Journal of Wuhan Polytechnic University, 2005, 24(2): 15-18 |
| |
| Authors: | HU Shu-jun |
| |
| Abstract: | Owing to the complex patters of the large remote sensing image, the common classifier based on SVMs produces large size of support vectors, which will decrease the detection efficiency. In this paper, a bottom up algorithm based on hierarchy of SVMs is used to improve the speed of detection, which is constructed by several SVMs classifiers whose complexity is gradually increased. The algorithm can perform efficiently by rapidly removing the obvious background and reducing the number of regions that are provided to detect by more complex classifiers. |
| |
| Keywords: | object detection support vector machines stratified detection. |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
