论文部分内容阅读
基于雷达一维距离像,将主成分分析方法和受限的库仑能量(Restricted Coulomb Energy,RCE)神经网络相结合,引入到雷达目标识别领域,以提高识别率。该方法先采用PCA方法提取导弹目标的主特征,然后利用RCE网络对导弹训练集样本进行学习,并将训练好的RCE网络作为分类器进行识别处理。试验结果表明,该方法具有良好的识别性能,能有效对导弹攻击目标予以判别。
Based on the radar one-dimensional range image, the principal component analysis method is combined with the restricted Coulomb energy (RCE) neural network to introduce radar target recognition to improve the recognition rate. In this method, PCA method is used to extract the main features of the missile target. Then, the RCE network is used to learn the missile training set samples, and the trained RCE network is used as a classifier to identify and process the missiles. The experimental results show that this method has good recognition performance and can effectively distinguish the target of missile attack.