Ge Hongyi, Jiang Yuying, Xu Zhaohui, Lian Feiyu, Zhang Yuan, Xia Shanhong
Opt Express. 2014 May 19;22(10):12533-44. doi: 10.1364/OE.22.012533.
The terahertz (THz) spectra in the range of 0.2-1.6 THz (6.6-52.8 cm) of wheat grains with various degrees of deterioration (normal, worm-eaten, moldy, and sprouting wheat grains) were investigated by terahertz time domain spectroscopy. Principal component analysis (PCA) was employed to extract feature data according to the cumulative contribution rates; the top four principal components were selected, and then a support vector machine (SVM) method was applied. Several selection kernels (linear, polynomial, and radial basis functions) were applied to identify the four types of wheat grain. The results showed that the materials were identified with an accuracy of nearly 95%. Furthermore, this approach was compared with others (principal component regression, partial least squares regression, and back-propagation neural networks). The comparisons showed that PCA-SVM outperformed the others and also indicated that the proposed method of THz technology combined with PCA-SVM is efficient and feasible for identifying wheat of different qualities.
利用太赫兹时域光谱技术研究了不同劣化程度（正常、虫蛀、发霉和发芽）小麦籽粒在0.2 - 1.6太赫兹（6.6 - 52.8厘米）范围内的太赫兹（THz）光谱。采用主成分分析（PCA）根据累积贡献率提取特征数据；选取前四个主成分，然后应用支持向量机（SVM）方法。应用几种选择核（线性、多项式和径向基函数）来识别这四种类型的小麦籽粒。结果表明，材料的识别准确率接近95%。此外，将该方法与其他方法（主成分回归、偏最小二乘回归和反向传播神经网络）进行了比较。比较结果表明，PCA - SVM的性能优于其他方法，同时也表明所提出的太赫兹技术与PCA - SVM相结合的方法对于识别不同品质的小麦是高效可行的。
