光谱学与光谱分析, 2013, 33 (6): 1537, 网络出版: 2013-06-07
近红外光谱结合主成分分析和BP神经网络的转基因大豆无损鉴别研究
Study on Near Infrared Spectroscopy of Transgenic Soybean Identification Based on Principal Component Analysis and Neural Network
摘要
为探究无损鉴别转基因大豆的可行性, 利用近红外光谱分析仪对大豆扫描得到反射光谱, 应用主成分分析结合BP神经网络方法进行分析鉴别。 首先应用主成分分析法, 得到包含大豆99.03%的光谱信息的6个主成分, 再将其作为BP神经网络的输入, 对应的大豆种类作为输出, 建立一个三层BP神经网络模型。 该模型对于转基因大豆的正确识别率为100%, 说明近红外光谱结合主成分分析和BP神经网络的方法能无损快速准确地鉴别转基因大豆。
Abstract
In order to explore a rapid identification method for transgenic soybeans, non-transgenic and transgenic soybeans were tested as the experimental samples via near infrared spectroscopy (NIR) and principal component analysis (PCA) combined with back propagation artificial neural network (BP-ANN) model. The spectrum data was collected after NIRS scanning the samples, and then analyzed by PCA plus BP-ANN model. The accumulative reliabilities of the six components were 99. 03% through the PCA. Then BP-ANN model was used to further test these six components and a three-layer BP-ANN model was developed. The final result achieved a 100% recognition rate of all 22 test samples respectively. In conclusion, the measure of NIRS and PCA combined with BP-ANN model has proved to be a rapid and accurate method to detect transgenic soybean nondestructively.
吴江, 黄富荣, 黄才欢, 张军, 陈星旦. 近红外光谱结合主成分分析和BP神经网络的转基因大豆无损鉴别研究[J]. 光谱学与光谱分析, 2013, 33(6): 1537. WU Jiang, HUANG Fu-rong, HUANG Cai-huan, ZHANG Jun, CHEN Xing-dan. Study on Near Infrared Spectroscopy of Transgenic Soybean Identification Based on Principal Component Analysis and Neural Network[J]. Spectroscopy and Spectral Analysis, 2013, 33(6): 1537.