离散小波特征提取及人工神经网络分类法的傅里叶变换红外光谱法识别鳞毛蕨科3种植物

利用水平衰减全反射-傅里叶变换红外光谱法测定了3种药用鳞毛蕨科植物贯众、阔鳞鳞毛蕨和变异鳞毛蕨根部的FT-IR.运用基于离散小波多分辨率分析法对FT-IR吸收较为相似的3种药用蕨类植物根的FT-IR进行特征提取.选择第4、5分解层数的特征向量,进行人工神经网络(Artificial neural network,ANN)训练;再用训练出来的网络对不同产地的3种药用蕨类植物根所得FT-IR小波提取的特征向量进行分类.通过对240个不同样本的预测,说明能够采用基于FT-IR-离散小波特征提取及人工神经网络分类法对同科3种药用蕨类植物根的FT-IR进行识别.

Recognition Among Three Kinds of Pteridophyte Plants Based on Fourier Transform Infrared-Discrete Wavelet Feature Extraction and Artificial Neural Network Classification Method

Fourier transform infrared(FT-IR) and horizontal attenuated total reflectance(HATR) techniques were used to obtain the FT-IR of three kinds of pteridophyte plants(the root of Cyrtomium fortunei J.Sm,Dryopteris championii(Bench) C.Chr.apud Ching and Dryopteris varia(L.) O.Ktze.).The similar features of FT-IR among the root of Cyrtomium fortunei J.Sm,Dryopteris championii(Bench) C.Chr.apud Ching and Dryopteris varia(L.) O.Ktze.were extracted by discrete wavelet transform.The scale 4 and 5 were used to extract the feature vectors,which were used to train the artificial neural network(ANN).The trained neural network was used to classify the root of Cyrtomium fortunei J.Sm,Dryopteris championii(Bench) C.Chr.apud Ching and Dryopteris varia(L.) O.Ktze.,which were collected from different places.According to 240 prediction samples,we could effectively identify the root of Cyrtomium fortunei J.Sm,Dryopteris championii(Bench) C.Chr.apud Ching and Dryopteris varia(L.) O.Ktze.by FT-IR with discrete wavelet feature extraction and artificial neural network classification.