基于人工神经网络的荧光光谱识别  

A neural network proposed to recognize the fluorescence spectrum

作　　者：范怀玉[1] 姚青[1] 申金媛[2] 

机构地区：[1]济宁医学院信息工程学院,山东日照276826 [2]郑州大学信息工程学院,河南郑州450001

出　　处：《济宁医学院学报》2010年第4期294-297,共4页Journal of Jining Medical University

基　　金：济宁医学院青年科研基金项目

摘　　要：目的利用人工神经网络的方法对杂质气体的荧光光谱进行识别。方法选用三层前馈型单输出的神经网络结构,学习算法选用自适应的RLS(Recursive Least Square)算法,在对基于RLS算法的三层前馈单输出的神经网络学习训练完成后,引入互联权重删剪算法对网络的输入层进行删剪。网络的结构不但得到了优化,而且计算复杂度也大大降低。提高网络的泛化能力;最后利用优化后的网络对测试样本进行识别。结果仿真实验表明,与删剪前的网络结构相比,在降低了网络的计算复杂度的同时,删剪优化后的正确识别率能够达到100%。结论利用RLS算法能够提高网络的运算速度,在此基础上的删剪算法能够降低计算的复杂度,并且能够对冗余信息进行了剔除,进而提高整个网络的识别能力。Objective A tapped delay neural network is proposed to recognize the fluorescence spectrum. Methods Firstly adaptive learning algorithm based on recursive least square is employed to train the tapped delay neural network,because this algorithm＇s learning step can be auto-conditioning and the number of it＇s tunable parameters is few,the convergence rate is fast.Secondly the architecture of neural network which has been trained is optimized by utilizing pruning algorithm to reduce the computational complexity and enhance network＇s generalization.And then the optimized network is retrained so that it has the optimum parameters.At last the test samples are predicted by the ultimate network.Results The simulation and comparison show that this optimized neuron network can not only reduce the calculating complexity greatly,but also the correct recognition rate is up to 100%.Conclusion The RLS algorithm can improve network calculation speed,and the pruning algorithm can not only reduce the computational complexity but also remove redundant information to enhance the recognition ability of the entire network.

关 键 词：荧光光谱 均方差 删剪算法 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]