为提高不同海拔茶叶品质近红外光谱技术鉴别方法的精度,提出采用局部线性嵌入法（LLE）和拉普拉斯特征映射法（LE）非线性流形学习方法对近红外光谱数据进行降维处理,并与基于核函数的非线性（KPCA）及线性（PCA）降维方法比较,建立不同海拔茶叶品质的近红外光谱LSSVM鉴别模型。不同降维方法可视化结果表明,KPCA和PCA方法的数据点离散性较大,400~800 m和800~1 200 m的样本点重叠较多,而非线性流形学习方法能将同一类样本点在三维空间很好地聚集在一起,不同海拔的茶叶能较好地区分开,且聚集效果方面LE方法好于LLE方法。模型性能表明,LE_LSSVM模型性能最佳,预测集总体判别率、Kappa系数分别为100%和1.00;相比于PCA_LSSVM、KPCA_LSSVM和LLE_LSSVM,模型预测集总体判别率分别提高1.7%、1.7%、3.3%;Kappa系数分别提高0.025、0.03、0.05。研究表明,LE等非线性流形学习降维方法在近红外光谱数据降维、简化模型复杂度、提高模型精度方面效果很好,为茶叶品质快速检测方法研究提供了一种新思路。
In order to improve the accuracy of near infrared (NIR) spectroscopy identification methods for tea at different altitude, the non-linear manifold dimensionality reduction methods of locally linear embedding (LLE) and laplacian eigenmaps (LE) were used to reduce the dimension of NIR spectral data, and compared with non-linear (KPCA) and linear (PCA) dimensional reduction methods. Discrimination models were established for tea at different altitude based on different dimensional reduction methods and least squares support vector machine (LSSVM) algorithm. Visualization of different dimensionality reduction results show that data processed by KPCA and PCA methods were more discrete. In particular, there were more overlaps between 400-800 m and 800-1 200 m samples. However, the same kind of sample points could be gathered well in three-dimensional space by the nonlinear manifold dimensionality reduction methods can. Tea at different altitude could be easily separated and the aggregation effect of the LE was better than that of the LLE. The results of models indicate the LE_LSSVM model had the best performance, with the prediction set accuracy and Kappa value of 100% and 1.00 respectively. Compared with PCA_LSSVM, KPCA_LSSVM and LLE_LSSVM models, the accuracy of prediction set was improved by 1.7%, 1.7%, 3.3% and Kappa values increased by 0.025, 0.03, and 0.05. The results show that LE and other nonlinear manifold dimensionality reduction methods were effective in reducing dimension of near infrared spectral data, simplifying model complexity, and improving model precision. The study provides a new means for rapid detecting for tea quality research.
[1] Han W, Huang J G, Li X, et al.Altitudinal effects on the quality of green tea in east China: a climate change perspective[J]. European Food Research and Technology, 2017, 243(2): 323-330.
[2] Zhuang X G, Wang L L, Chen Q, et al.Identification of green tea origins by near-infrared (NIR) spectroscopy and different regression tools[J]. Science China Technological Sciences, 2017, 60(1): 84-90.
[3] 陈美丽, 张俊, 龚淑英, 等. 茉莉花茶主要品质成分定量近红外光谱分析模型的建立[J]. 茶叶科学, 2013, 33(1): 21-26.
[4] 刘洋, 余天星, 李明玺, 等. 基于近红外光谱技术的信阳毛尖品质判别研究[J]. 现代食品科技, 2018, 34(8): 1-7.
[5] Ouyang Q, Liu Y, Chen Q S, et al.Intelligent evaluation of color sensory quality of black tea by visible-near infrared spectroscopy technology: A comparison of spectra and color data information[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2017, 180: 91-96. DOI: 10.1016/j.saa.2017.03.009.
[6] Jiang H, Chen Q S.Chemometric models for the quantitative descriptive sensory properties of green tea (Camellia sinensis L.) using fourier transform near infrared (FT-NIR) spectroscopy[J]. Food Analytical Methods, 2015, 8(4): 954-962.
[7] 王胜鹏, 龚自明, 高士伟, 等. 基于近红外光谱技术的恩施玉露茶保存年份的快速无损鉴别[J]. 华中农业大学学报, 2015, 34(5): 111-114.
[8] Shan R F, Cai W S, Shao X G.Variable selection based on locally linear embedding mapping for near-infrared spectral analysis[J]. Chemometrics and Intelligent Laboratory Systems, 2014, 131: 31-36. DOI: 10.1016/j.chemolab.2013.12.002.
[9] 林萍, 陈永明, 邹志勇. 非线性流形降维与近红外光谱分析技术的大米贮藏期快速判别[J]. 光谱学与光谱分析, 2016, 36(10): 3169-3173.
[10] 李庆波, 贾召会. 一种光谱分析中的降维方法[J]. 光谱学与光谱分析, 2013, 33(3): 780-784.
[11] 黄宏臣, 张倩倩, 韩振南, 等. 拉普拉斯特征映射算法在滚动轴承故障识别中的应用[J]. 中国测试, 2015, 41(5): 94-98.
[12] Zhang Y, Ye D, Liu Y.Robust locally linear embedding algorithm for machinery fault diagnosis[J]. Neurocomputing, 2018, 273: 323-332.
[13] 金瑞, 李小昱, 颜伊芸, 等. 基于高光谱图像和光谱信息融合的马铃薯多指标检测方法[J]. 农业工程学报, 2015, 31(16): 258-263.
[14] 孙伟伟, 刘春, 李巍岳. 联合改进拉普拉斯特征映射和k-近邻分类器的高光谱影像分类[J]. 武汉大学学报(信息科学版), 2015, 40(9): 1151-1156.
[15] 张赟, 杨栋, 斯彦刚, 等. 基于监督流形学习的航空发动机振动故障诊断方法[J]. 推进技术, 2017, 38(5): 1147-1154.
[16] 钱进, 邓喀中, 范洪冬. 基于拉普拉斯特征映射高光谱遥感影像降维及其分类[J]. 遥感信息, 2012, 27(5): 3-7.
[17] Mantziou E, Papadopoulos S, Kompatsiaris Y.Learning to detect concepts with approximate laplacian eigenmaps in large-scale and online settings[J]. International Journal of Multimedia Information Retrieval, 2015, 4(2): 95-111.
[18] Singer A, Wu H.Spectral convergence of the connection Laplacian from random samples[J]. Information and Inference: A Journal of the IMA, 2016, 6(1): 58-123.
[19] 吴尚蓉, 陈仲新, 任建强, 等. 定位尺度和像元空间关系对GF-1亚像元定位精度影响分析[J]. 农业工程学报, 2016, 32(5): 163-171.
[20] 王冰玉, 孙威江, 黄艳, 等. 基于遗传算法的安溪铁观音品质快速评价研究[J]. 光谱学与光谱分析, 2017, 37(4): 1100-1104.
浙公网安备 33019902000101号 
