小波神经网络在农药荧光光谱识别中的应用

Application of Wavelet Neural Network to the Fluorescence Spectral Recognition of the Pesticides

对于结构非常相似的农药,它们的荧光光谱也非常相似并且在很宽波长范围内相互重叠。传统的荧光光谱分析法很难对其进行分类识别。一种基于小波分析而构造的新型神经网络———小波神经网络是利用它并适当选取网络结构和小波基,实现了对卡死克、盖虫散和吡虫啉三种农药荧光光谱的分类识别。实验表明,小波神经网络对光谱间的细微结构差别具有良好的识别能力。通过比较发现,在分类识别方面小波神经网络比BP网络具有更高的分辨率及较少的训练次数。

For the pesticides with similar structures, their fluorescence spectra are also similar and overlapped in a wide wavelength arrange. The conventional fluorescence spectrum analysis method can hardly identify them. A new type of neural network wavelet neural network is introduced, which is constructed based on wavelet analysis. The classification of flufenoxuron, hexaflumuron and imidacloprid are realized with adaptive network structure and wavelet basis. The experiment results show that wavelet neural network has the better ability to the fine structure difference between the spectra. Compared with BP networks, wavelet neural network has higher resolution and less training times.