Please wait a minute...
J4  2010, Vol. 44 Issue (3): 619-624    DOI: 10.3785/j.issn.1008-973X.2010.03.036
环境工程、食品工程     
应用有效波长进行奶茶品种鉴别的研究
刘飞, 王莉, 何勇
浙江大学 生物系统工程与食品科学学院,浙江 杭州 310029
Application of effective wavelengths for variety identification
of instant milk teas
LIU Fei, WANG Li, HE Yong
College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310029, China
 全文: PDF  HTML
摘要:

基于最小二乘支持向量机建模方法,提出应用奶茶在可见/近红外光谱谱区的有效波长进行其品种鉴别的新方法.用225个样本建模,75个样本进行预测.通过对光谱数据进行偏最小二乘法分析,根据载荷图和回归系数图选择鉴别奶茶品种的有效波长(EW),并建立EW与最小二乘支持向量机(LSSVM)相结合的EWLSSVM模型,同时与应用主成分(PC)和小波变换(WT)建立的PCLSSVM和WTLSSVM模型进行判别准确率的比较.结果表明,应用EW、PC和WT建立的模型对建模样本的判别准确率均为100%,对预测集样本判别准确率分别为98.7%、98.7%和100%,获得了理想的鉴别效果.研究表明,应用可见/近红外光谱谱区的有效波长进行奶茶品种鉴别是可行的,且EWLSSVM模型能获得满意的鉴别精度.

Abstract:

Based on least squaressupport vector machine (LSSVM), the effective wavelength (EW) in visible/near infrared (Vis/NIR) region was proposed as a new approach for the variety discrimination of instant milk teas. This method uses 225 milk tea samples for the calibration set, while 75 samples for the validation set. After partial least squares (PLS) analysis, the EWs were selected according to the Xloading weights and regression coefficients, and an EWLSSVM model was developed for the variety discrimination. The PCLSSVM model using principal components (PCs) and the WTLSSVM model using wavelet transform (WT) were built for comparison. The recognition ratios of calibrations using EW, PC and WT were all 100%, for the calibration set, while 98.7%, 98.7% and 100% for the validation set, respectively. An excellent recognition ratio was achieved by these three models. It is feasible to use effective wavelengths in Vis/NIR region for the variety discrimination of instant milk teas and the EWLSSVM model can achieve a satisfying recognition ratio.

出版日期: 2012-03-20
:  O 657.33  
基金资助:

 国家自然科学基金资助项目(30671213);教育部高等学校优秀青年教师教学科研奖励计划资助项目(02411);浙江省自然科学基金资助项目(Y307158);中央高校基本科研业务费专项资金资助项目.

通讯作者: 何勇,男,教授.     E-mail: yhe@zju.edu.cn
作者简介: 刘飞(1983—),男,河北定州人,博士生,主要从事数字农业和农产品多光谱检测技术研究.Email: fliu@zju.edu.cn
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

刘飞, 王莉, 何勇. 应用有效波长进行奶茶品种鉴别的研究[J]. J4, 2010, 44(3): 619-624.

LIU Fei, WANG Chi, HE Yong. Application of effective wavelengths for variety identification
of instant milk teas. J4, 2010, 44(3): 619-624.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2010.03.036        http://www.zjujournals.com/eng/CN/Y2010/V44/I3/619

[1] 严衍禄,赵龙莲,韩东海,等.近红外光谱分析基础与应用[M].北京:中国轻工业出版社,2004.
[2] 李晓丽,胡兴越,何勇.基于主成分和多类判别分析的可见近红外光谱水蜜桃品种鉴别新方法[J].红外与毫米波学报,2006,25(6): 417420.
LI Xiaoli, HU Xingyue, HE Yong. New approach of discrimination of varieties of juicy peach by near infrared spectra based on PCA and MDA model [J]. Journal of Infrared Millimeter Waves, 2006, 25(6): 417420.
[3] 何勇,李晓丽.近红外光谱杨梅品种鉴别方法的研究[J].红外与毫米波学报,2006,25(3): 192194.
HE Yong, LI Xiaoli. Discrimination of varieties of waxberry using near infrared spectra [J]. Journal of Infrared Millimeter Waves, 2006, 25(3): 192194.
[4] 刘燕德,应义斌,傅霞萍,等.一种近红外光谱水果内部品质自动检测系统[J].浙江大学学报:工学版,2006,40(1): 5356.
LIU Yande, YING Yibin, FU Xiaping, et al. Atomatic measurement system of fruit internal quality using nearinfrared spectroscopy [J]. Journal of Zhejiang University: Engineering Science, 2006, 40(1): 5356.
[5] 刘燕德,应义斌.基于MATLAB语言的苹果糖度近红外光谱定量分析[J].浙江大学学报:工学版,2004,38(10): 13711374.
LIU Yande, YING Yibin. Quantitative analysis of near infrared spectra in apple sugar content based on MATLAB[J]. Journal of Zhejiang University: Engineering Science, 2004, 38(10): 13711374.
[6] LIU F, HE Y, WANG L, et al. Feasibility of the use of visible and near infrared spectroscopy to assess soluble solids content and pH of rice wines [J]. Journal of Food Engineering, 2007, 83: 430435.
[7] 刘飞,何勇,王莉.黄酒糖度预测的可见近红外光谱方法研究[J].光学学报,2007,27(11): 20542058.
LIU Fei, HE Yong, WANG Li. Methods for the prediction of sugar content of rice wine using visiblenear infrared spectroscopy[J]. Acta Optica Sinica, 2007, 27(11): 20542058.
[8] WU D, HE Y, FENG S J, et al. Study on infrared spectroscopy technique for fas measurement of protein content in milk powder based on LSSVM [J]. Journal of Food Engineering, 2008, 84: 124131.
[9] BORIN A, FERRO M F, MELLO C, et al. Leastsquares support vector machines and near infrared spectroscopy for quantification of common adulterants in powdered milk [J]. Analytica Chimica Acta, 2006, 579: 2532.
[10] CHEN Q S, ZHAO J W, FANG C H, et al. Feasibility study on identification of green, black and Oolong teas using nearinfrared reflectance spectroscopy based on support vector machine (SVM) [J]. Spectrochimica Acta Part A, 2007, 66: 568574.
[11] LI X L, HE Y, WU C Q, et al. Nondestructive measurement and fingerprint analysis of soluble solid content of tea soft drink based on Vis/NIR spectroscopy [J]. Journal of Food Engineering, 2007, 82: 316323.
[12] 白锁柱,陈保国,张力,等.乳化法——火焰原子吸收光谱法测定奶茶粉中铁[J].理化检验化学分册,2004,40(11): 648649.
BAI Suozhu, CHEN Baoguo, ZHANG Li, et al. FAAS determination of iron in milk tea powder\
[J\]. PTCA Part B: Chemical Analysis, 2004, 40(11): 648649.
[13] FERRUZZI M G, GREEN R J. Analysis of catechins from milktea beverages by enzyme assisted extraction followed by high performance liquid chromatography [J]. Food Chemistry, 2006, 99: 484491.
[14] 褚小立,袁洪福,陆婉珍.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4): 528539.
CHU Xiaoli, YUAN Hongfu, LU Wanzhen. Progress and application of spectral data pretreatment and wavelength selection methods in NIR analytical technique [J]. Progress in Chemistry, 2004, 16(4): 528539.
[15] VAPNIK V N. The nature of statistical learning theory [M]. New York: SpringerVerlag, 1995.
[16] 朱家元,杨云,张恒喜,等.基于优化最小二乘支持向量机的小样本预测研究[J].航空学报,2004,25(6): 565568.
ZHU Jiayuan, YANG Yun, ZHANG Hengxi, et al. Data prediction with few observations based on optimized least squares support vector machines [J]. Acta Aeronautica et Astronautica Sinica, 2004, 25(6): 565568.

No related articles found!