摘要
提出了一种基于模糊神经网络(Fuzzy Neural Networks,FNN)的目标识别算法。在对目标进行模糊化处理基础上,通过主成分分析(Principal Component Analysis,PCA)提取相应的特征空间,以畸变的特征向量对系统进行训练,从而获得较高的目标变化适应性。仿真实验结果表明,这种方法具有较强的自适应能力,抗噪性能也有所提高。
This paper present a effective method for target identification based on Fuzzy Neural Network (FNN). Firstly, the target images are processed for fuzzification and distortion. Secondly, the eigenspace is obtained based on Principal Component Analysis (PCA). Lastly, the networks is trained with the distorted eigenvectors for higher adaptability of target variation. The results of simulation show that the correctness of this method is higher than that of networks with conventional training method. Obviously,the ability to counteract disturbance and noise is also raised.
出处
《火力与指挥控制》
CSCD
北大核心
2008年第10期24-26,33,共4页
Fire Control & Command Control
关键词
目标识别
模糊神经网络
主成分分析
target recognition,fuzzy neural networks,principal component analysis