论文部分内容阅读
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 high-resolution radar target one-dimensional distance image recognition algorithm. The algorithm borrows the image processing method, first encodes the one-dimensional distance image of radar target, and then extracts a set of shape features through Fourier transform. This set of features accurately describes the course of a one-dimensional distance profile. Then, the artificial neural network technology is used to identify the group of shape features within a certain range of attitude angles. Experimental results show that good recognition can be obtained. The algorithm provides an effective way for the real-time processing of radar target one-dimensional distance image recognition algorithm.