摘要
文中提出了一种高分辨率雷达目标一维距离像的识别算法。该算法借用图像处理方法，先对雷达目标的一维距离像进行编码，经过傅里叶变换提取一组形状特征。这组特征精确地描述了一维距离像曲线的走向。而后，利用人工神经网络技术对一定姿态角变化范围内的这组形状特征进行识别。实验结果表明，可以获得良好的识别效果。该算法为雷达目标一维距离像识别算法的实时处理提供了一条有效途径。
In this paper, a new method of recognizing one-dimensional range profiles of high resolution radar targets is proposed. The algorithms of image processing are used in this method. First, we code one-dimensional range profiles, then, use Fourier transformation to extract a series of formal features which can be recognized by neural networks. The experimental results show that, in some aspect angle interval .the recognition is good.
出处
《国防科技大学学报》
EI
CAS
CSCD
北大核心
1996年第4期62-65,87,共5页
Journal of National University of Defense Technology
基金
国防预研基金
关键词
雷达
一维距离像
图像处理
分辨率
目标识别
one-dimensional range profiles of radar targets, image processing algorithms, neural networks, targets recognition
