论文部分内容阅读
针对神经网络及其改进的神经网络易陷入局部极小值、精度和泛化之间不可调和等固有缺陷,本文将主成分分析(PCA)优化径向基函数神经网络(RBFN)应用于毫米波辐射计的目标识别,利用该方法对探测目标较多的情况下进行目标识别,并与BP和传统RBFN神经网络方法进行比较。结果表明:经过主成分分析优化的径向基函数神经网络相比BP神经网络和传统RBFN神经网络对目标的预测精度更高、发生错判的几率更低、识别效果更好。
Aiming at the inherent defects such as the local minimum of neural network and its improved neural network, and the irreconcilability between accuracy and generalization, this paper applies principal component analysis (RBFN) to Radial Basis Function Neural Network (RBFN) Radiometer target recognition, this method is used to target detection in the case of more targets, and compared with BP and traditional RBFN neural network method. The results show that the radial basis function neural network optimized by principal component analysis has higher prediction accuracy than BP neural network and traditional RBFN neural network, the probability of misjudgment is lower and the recognition effect is better.