摘要
合成孔径雷达自动目标识别是目前国内外模式识别领域的重点研究课题之一。本文给出了一种内存需求小,低计算复杂度且具有较好识别性能的SAR图像目标识别方法,先通过自适应阈值分割来获得目标图像,然后提取其中心矩特征,采用SVM来进行识别。基于美国M STAR实测数据的识别试验验证了该方法的有效性。
SAR images based radar automatic target recognition (ATR) is a hot topic in pattern recognition community. This paper proposes a SAR ATR method featuring low memory requirement, low computation complexity and good recognition performance. The target SAR images are extracted via adaptive threshold image segmentation algorithm, and its central moment features are extracted and classified by support vector machine (SVM) classifier. The efficiency of the proposed methods is evaluated based on the MSTAR data.
出处
《火控雷达技术》
2006年第2期74-77,共4页
Fire Control Radar Technology
关键词
合成孔径雷达
自动目标识别
中心矩特征
SVM分类器
synthetic aperture radar
automatic target recognition
central moment feature, SVM classifier