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。
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.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第9期212-214,共3页
Computer Engineering and Applications
关键词
支持向量机
遥感
目标检测
样本选取
Support Vector Machines,remote sensing,object detection,samples selection
