基于径向基神经网络的浮游植物分类系统

SPECTRA OF PHYTOPLANKTON CLASSIFICATION SYSTEM BASED ON RADIAL BASIS NEURAL NETWORK

应用模糊聚类和小波变换提取浮游植物活体的特征光谱,并以此为输入向量,引入径向基函数网络对浮游植物的光谱进行分类识别,建立了适用于光谱识别的径向基函数神经网络系统。结果表明,该方法较传统的统计方法更方便,识别准确率更高。

A radius basis function （RBF） network was introduced to realize the classification and the recognition of spectrum of phytoplankton, based on the extraction of characteristic spectrum using fuzzy clustering and wavelet transformation. Thus a neural network system based on RBF network adapting the spectrum recognition was established by using those characteristic parameters as the input vectors. It was found that the neural network based on RBF was superior to conventional statistic method in convenience and recognition accuracy.