中国农业科学 ›› 2015, Vol. 48 ›› Issue (20): 4111-4119.doi: 10.3864/j.issn.0578-1752.2015.20.012

• 园艺·贮藏·保鲜·加工 • 上一篇    下一篇

基于近红外光谱的纽荷尔脐橙产地识别研究

廖秋红1,2,何绍兰1,谢让金1,钱春2,胡德玉1,2,吕强1,易时来1,郑永强1,邓烈1

 
  

  1. 1中国农业科学院柑桔研究所/西南大学柑桔研究所/国家柑桔工程技术研究中心,重庆 400712
    2西南大学园艺园林学院,重庆 400715
  • 收稿日期:2015-03-15 出版日期:2015-10-20 发布日期:2015-10-20
  • 通讯作者: 邓烈,E-mail:liedeng@cric.cn E-mail:liedeng@cric.cn
  • 作者简介:廖秋红,E-mail:qiu5711@163.com
  • 基金资助:
    国家“863”计划(2012AA101904)、国家国际科技合作专项(2013DFA11470)、重庆市科技支撑项目(cstc2014fazktpt80015)

Study on Producing Area Classification of Newhall Navel Orange Based on the Near Infrared Spectroscopy

LIAO Qiu-hong1,2, HE Shao-lan1, XIE Rang-jin1, QIAN Chun2, HU De-yu1,2, LÜ Qiang1, YI Shi-lai1, ZHENG Yong-qiang1, DENG Lie1   

  1. 1Citrus Research Institute, Southwest University/Chinese Academy of Agricultural Sciences/ National Engineering Technology Research Center for Citrus, Chongqing 400712
    2College of Horticulture and Landscape Architecture, Southwest University, Chongqing 400715
  • Received:2015-03-15 Online:2015-10-20 Published:2015-10-20

摘要: 【目的】中国柑橘产区分布广、生态类型复杂,不同产地纽荷尔脐橙果实品质和市场效应具有较大差异。研究基于近红外光谱技术的柑橘产地识别技术,利于不同柑橘产地果品的识别和鉴伪。【方法】从中国南方17个纽荷尔脐橙主要产地选择代表性成年果园,分别采摘成熟鲜果样品100个。利用SupNIR-1500近红外分析仪采集脐橙果实赤道部、肩部表面以及果汁滤液的近红外反射光谱,光谱波长范围为1 000—2 499 nm。采用主成分分析法对原始光谱数据进行预处理,提取近红外光谱的特征信息以降低数据集维度以及噪声。研究人工神经网络理论,构建由一个输入层、一个具有非线性激励函数的隐藏层和一个输出层组成的典型的3层人工神经网络识别模型。研究由径向基函数作为核函数、以光谱主成分作为输入的支持向量机模型,构建由126个分类器组成的一对一扩展支持向量机模型。研究遗传算法优异的自然选择特性,利用遗传算法从光谱主成分中选择出最优的特征基因子集作为支持向量机的输入,构建遗传算法-支持向量机模型。利用3种模型分别对果汁滤液的近红外反射光谱数据进行分类,从而实现产地识别测试,并根据产地识别精度筛选出最优的产地识别模型。进一步对比该最优识别模型对果实赤道部、肩部反射光谱数据的识别精度,从而确定识别精度最高的光谱数据采集源。【结果】利用所建立的3层人工神经网络模型对纽荷尔脐橙果汁滤液的近红外光谱进行产地识别测试,确定当输入神经元数量为11、隐藏神经元数量为13时,模型对果实产地识别的最佳精度达81.45%。采用一对一扩展方式建立支持向量机产地识别模型,研究确定采用径向基函数作为核函数,当主成分数量为20时,脐橙产地识别精度最高可达86.98%。测试利用遗传算法-支持向量机混合模型进行脐橙产地分类识别,确定当种群数量为200、遗传代数为100、交叉概率0.7、突变概率0.01时,遗传算法选择出最优的基因子集进行产地识别,遗传算法-支持向量机模型的产地识别精度最高可达89.72%,优于人工神经网络分类模型和支持向量机分类模型的产地识别精度。进一步利用遗传算法-支持向量机产地识别模型对果实赤道部及肩部的果面反射光谱进行产地识别测试,得到对应的最高识别精度分别为80.00%和69.00%。【结论】遗传算法-支持向量机模型对果汁反射近红外光谱进行产地识别精度最高,优于人工神经网络模型和支持向量机模型。该模型对果实赤道部反射光谱进行分类的精度次于果汁滤液反射光谱但优于果实肩部反射光谱,因此,可利用赤道部的反射光谱实现非破坏性果实产地分类识别

关键词: 纽荷尔脐橙, 产地识别, 近红外光谱, 主成分分析, 人工神经网络, 支持向量机, 遗传算法

Abstract: 【Objective】Newhall navel orange (Citrus sinensis L.) fruits from different producing areas in China, exhibit contrasting quality and market values, due to wide-spreading location of various ecologies. Developing a recognition method based on Near-Infrared (NIR) spectroscopy is very important to identify and distinguish fruits from different producing areas. 【Method】One representative orchard was selected from 17 main producing areas distributed throughout southern China, from which one 100 Newhall navel orange samples were collected. NIR spectra were collected with a SupNIR-1500 spectrograph from the surface of fruit equator and shoulder and the filtered juice for each sample, which were further preprocessed through principal component analysis (PCA) for reduced dimensions and noise. By studying artificial neural network (ANN), a classic three-layer ANN model was established with an input layer, a hidden layer of non-linear activation function and an output layer. By studying support vector machine (SVM) with the radial basis function (RBF) being the kernel function and the principal components of NIR spectra being the input, a one-to-one extended SVM model was established with 126 classifiers. Gene algorithm (GA) with excellent natural selection was used to identify the best Genetic character subset from the principal components as inputs of a SVM classifier, thus a GA-optimized SVM model was composed. These three models were used to classify the NIR spectra of filtered juice, thus the production areas of the oranges, the classification accuracies of which decided the best classifier. Furthermore, the accuracies of the best classifier were tested with the NIR spectra from fruit equator and shoulder surface being the inputs. As a comparison, the best NIR spectra could be identified. 【Result】 Producing area classification was implemented with the three-layer ANN classifier with NIR spectra of Newhall orange juice as the input, where the classifying accuracy reached up to 81.45% when there were 11 input neurons and 13 hidden neurons. The studied one-to-one extended SVM classifier with radial basis function being the core function, exhibited higher accuracy of 86.98% when the number of PC was 20, better than ANN classifier. For GA-SVM classifier took into account the interaction of individual inputs, where the PCA-processed results were optimized by GA. During the experiments, classification accuracy hit 89.72% when the population, generation, mating probability, and mutation probability were 200, 100, 0.7 and 0.01 respectively, surpassed ANN and SVM classifier. Subsequent research found the highest accuracy of GA-SVM classifier was 80% when taking the spectra from the fruit equator, and 69% from the shoulder, not good enough as that of orange juice.【Conclusion】Considering the accuracy, GA-SVM classifier was regarded with the most excellence among three investigated classifiers. Spectra of orange juice were selected as the best data to analyze origins traceability. Accuracy of spectra of fruit equator was inferior to juice but superior to the shoulder, thus had the potential for non-destructive origins classification.

Key words: Newhall Navel orange, producing area recognition, near-infrared spectroscopy, principal components analysis, artificial neural network, support vector machine, genetic algorithm

[1]    鲍江峰, 夏仁学, 彭抒昂, 李国怀. 湖北省纽荷尔脐橙园土壤营养状况及其对果实品质的影响. 土壤, 2006, 38(1): 75-80.
Bao J F, Xia R X, Peng S A, Li G H. Soil nutrient status of Newhall orchards of Hubei province and its effect on fruit quality of Newhall orange. Soils, 2006, 38 (1): 75-80. (in Chinese)
[2]    樊卫国, 王立新. 不同供磷水平对纽荷尔脐橙幼树生长及叶片营养元素含量的影响. 中国农业科学, 2012, 45(4): 714-725.
Fan W G, Wang L X. Effect of different phosphorus levels on growth and leaf element contents of Young Newhall Navel orange trees. Scientia Agricultura Sinica, 2012, 45(4): 714-725. (in Chinese)
[3]    韩春丽. 弱光下纽荷尔脐橙光合生理特性研究[D]. 武汉: 华中农业大学, 2008.
Han C L. Studies on the photosynthetic physiologieal characteristics of Cittrus sinensis cv. Newhalll eaves under low light stress [D]. Wuhan: Huazhong Agricultural University, 2008. (in Chinese)
[4]    Simpkins W A, Louie H, Wu M, Harrison M, Goldberg D. Trace elements in Australian orange juice and other products. Food Chemistry, 2000, 71(4): 423-433.
[5]    Rummel S, Hoelzl S, Horn P, Rossmann A, Claus S. The combination of stable isotope abundance ratios of H, C, N and S with 87Sr/86Sr for geographical origin assignment of orange juices. Food Chemistry, 2010, 118(4): 890-900.
[6]    Le Nguyen D D, Gemrot E, Loiseau G, Montet D. Determination of citrus fruit origin by using 16S rDNA fingerprinting of bacterial communities by PCR-DGGE: an application to clementine from Morocco and Spain. Fruits, 2008, 63(02): 75-84.
[7]    苏学素, 张晓焱, 焦必宁, 曹维荃. 基于近红外光谱的脐橙产地溯源研究. 农业工程学报, 2012, 28(15): 240-245.
Su X S, Zhang X Y, Jiao B N, Cao W Q. Determination of geographical origin of navel orange by near-infrared spectroscopy. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(15): 240-245. (in Chinese)
[8]    徐广通, 袁洪福, 陆婉珍. 现代近红外光谱技术及应用进展. 光谱学与光谱分析, 2000, 20(2): 134-142.
Xu G T , Yuan H F, Lu W Z. Development of modern near infrared spectroscopic techniques and its applications. Spectroscopy and Spectral Analysis, 2000, 20(2): 134-142. (in Chinese)
[9]    严衍禄, 赵龙莲. 现代近红外光谱分析的信息处理技术. 光谱学与光谱分析, 2000, 20(6): 777-780.
Yan Y L, Zhao L L. Information technology of modern NIR spectral analysis. Spectroscopy and Spectral Analysis, 2000, 20(6): 777-780. (in Chinese)
[10]   蔡先峰, 郭波莉, 魏益民, 孙淑敏, 赵多勇, 魏帅. 牛肉近红外光谱的地域及饲养期特征分析. 中国农业科学, 2011, 44(20): 4272-4278.
Cai X F, Guo B L, Wei Y M, Sun S M, Zhao D Y, Wei S. Analysis on characteristics of near infrared spectra of beef according to regions and feeding periods. Scientia Agricultura Sinica, 2011, 44(20): 4272-4278. (in Chinese)
[11]   张宁, 张德权, 李淑荣, 李庆鹏. 近红外光谱结合 SIMCA法溯源羊肉产地的初步研究. 农业工程学报, 2008, 24(12): 309-312.
Zhang N, Zhang D Q, Li S R, Li Q P. Preliminary study on origin traceability of mutton by near infrared reflectance spectroscopy coupled with SIMCA method. Transactions of the Chinese Society of Agricultural Engineering, 2008, 24(12): 309-312. (in Chinese)
[12]   李水芳, 单杨, 朱向荣, 李忠海. 近红外光谱结合化学计量学方法检测蜂蜜产地. 农业工程学报, 2011, 27(8): 350-354.
Li S F, Shan Y, Zhu X R, Li Z H. Detection of geographical origin of honey using near-infrared spectroscopy and chemometrics. Transactions of the Chinese Society of Agricultural Engineering, 2011, 27(8): 350-354. (in Chinese)
[13]   汤丽华, 刘敦华. 基于近红外光谱技术的枸杞产地溯源研究. 食品科学, 2012, 32(22): 175-178.
Tang L H,Liu D H. Tracing the geographic origin of Chinese wolfberry by Near-Infrared spectroscopy. Food Science, 2012, 32(22): 175-178. (in Chinese)
[14]   赵海燕, 郭波莉, 魏益民, 孙波, 孙淑敏, 严军辉, 张磊. 近红外光谱对小麦产地来源的判别分析. 中国农业科学, 2011, 44(7): 1451-1456.
Zhao H Y, Guo B L, Wei Y M, Sun B, Sun S M, Yan J H, Zhang L. Identification of geographical origins of wheat with discriminant analysis by Near-Infrared spectroscopy. Scientia Agricultura Sinica, 2011, 44(7): 1451-1456. (in Chinese)
[15]   Cozzolino D, Smyth H E, Gishen M. Feasibility study on the use of visible and near-infrared spectroscopy together with chemometrics to discriminate between commercial white wines of different varietal origins. Journal of Agricultural and Food Chemistry, 2003, 51: 7703-7708.
[16]   陈永明, 林萍, 何勇. 基于遗传算法的近红外光谱橄榄油产地鉴别方法研究. 光谱学与光谱分析, 2009, 29(3): 671-674.
Chen Y M, Lin P, He Y. Study on discrimination of producing area of olive oil using Near-Infrared spectra based on genetic algorithms. Spectroscopy and Spectral Analysis, 2009, 29(3): 671-674. (in Chinese)
[17]   Abdi H, Williams L J. Principal component analysis. Wiley Interdisciplinary Reviews: Computational Statistics, 2010, 2: 433-459.
[18]   何勇, 李晓丽, 邵咏妮. 基于主成分分析和神经网络的近红外光谱苹果品种鉴别方法研究. 光谱学与光谱分析, 2006, 26(5): 850-853.
He Y, Li X L, Shao Y N. Discrimination of varieties of apple using Near-Infrared spectra based on principal component analysis and artificial neural network model. Spectroscopy and Spectral Analysis, 2006, 26(5): 850-853. (in Chinese)
[19]   Dettmar H P, Barbour G S, Blackwell K T, Vogl T P, Alkon D L, Fry Jr F S, Totah J E, Chambers T L. Orange juice classification with a biologically based neural network. Computers and Chemistry, 1996, 20(2): 261-266.
[20]   Hecht-Nielsen R. Theory of the backpropagation neural network. International Joint Conference on Neural Networks, 1989: 593-605.
[21]   陈建, 陈晓, 李伟, 王加华, 韩东海. 基于近红外光谱技术和人工神经网络的玉米品种鉴别方法研究. 光谱学与光谱分析, 2008, 28(8): 1806-1809.
Chen J, Chen X, Li W, Wang J H, Han D H. Study on discrimination of corn seed based on near-infrared spectra and artificial neural network model. Spectrscopy and Spectral Analysis, 2008, 28(8): 1806-1809. (in Chinese)
[22]   Vapnik V, Golowich S E, Smola A. Support vector method for function approximation, regression estimation, and signal processing. Advances in Neural Information Processing Systems, 1997: 281-287.
[23]   Hidalgo J I, Matias P M, Lanchares J, Baraglia R, Tirado F, Garnica O. Hybrid parallelization of a compact genetic algorithm. 16th Euromicro Conference on Parallel, Distributed and Network-Based Processing, IEEE, 2003: 449-455.
[24]   Blanz V, Schölkopf B, Bülthoff H, Burges C, Vapnik V, Vetter T. Comparison of view-based object recognition algorithms using realistic 3D models. In Artificial Neural Networks—ICANN 96, Springer Berlin Heidelberg, 1996: 51-256.
[25]   丁建立, 陈增强, 袁著祉. 遗传算法与蚂蚁算法的融合. 计算机研究与发展, 2003, 40(9): 1351-1356.
Ding J L, Chen Z Q, Yuan Z Z. On the combination of genetic algorithm and ant algorithm. Journal of Computer Research and Development, 2003, 40(9): 1351-1356. (in Chinese)
[26] 中国气象科学数据共享服务网. http://cdc.cma.gov.cn/home.do.
Meteorological data sharing service network of Chine. http://cdc. cma.gov.cn/home.do. (in Chinese)
[1] 李文涛,杨江波,张绩,王克健,邓烈,吕强,何绍兰,谢让金,郑永强,马岩岩,易时来. 基于不同传感器的纽荷尔脐橙叶片叶绿素含量检测技术评价[J]. 中国农业科学, 2018, 51(6): 1057-1066.
[2] 李祥栋,潘虹,陆秀娟,魏心元,陆平,石明,秦礼康. 薏苡种质的主要营养组分特征及综合评价[J]. 中国农业科学, 2018, 51(5): 835-842.
[3] 陈雪萍,刘世尧,尹能文,荆凌云,魏丽娟,林呐,肖阳,徐新福,李加纳,刘列钊. 甘蓝型油菜茎秆纤维组分含量和木质素单体G/S 近红外模型构建[J]. 中国农业科学, 2018, 51(4): 688-696.
[4] 杜斌,胡笑涛,王文娥,马黎华,周始威. 交替沟灌玉米灌浆期茎流影响因子敏感性分析与模型适用性研究[J]. 中国农业科学, 2018, 51(2): 233-245.
[5] 苏杭,王琦,李春梅,邵华,金茂俊,王珊珊,郑鹭飞,佘永新,王静,王怀松,金芬. 植物生长调节剂噻苯隆对甜瓜品质的影响[J]. 中国农业科学, 2018, 51(16): 3095-3105.
[6] 张永玲,姜梦洲,俞佩仕,姚青,杨保军,唐健. 基于多特征融合和稀疏表示的农业害虫图像识别方法[J]. 中国农业科学, 2018, 51(11): 2084-2093.
[7] 栗振义,张绮芯,仝宗永,李跃,徐洪雨,万修福,毕舒贻,曹婧,何峰,万里强,李向林. 不同紫花苜蓿品种对低磷环境的形态与生理响应分析[J]. 中国农业科学, 2017, 50(20): 3898-3907.
[8] 邵小龙,汪楠,时小转,沈飞,宋伟,张强. 水稻生长过程中籽粒水分状态和横向弛豫特性分析[J]. 中国农业科学, 2017, 50(2): 240-249.
[9] 宋伟,胡婉君,徐宗季,刘兵,樊艳,倪阳,杨慧萍. 湖南省主要早籼稻品种挥发性物质种类及含量[J]. 中国农业科学, 2017, 50(2): 348-361.
[10] 路之娟,张永清,张楚,刘丽琴,杨春婷. 不同基因型苦荞苗期抗旱性综合评价及指标筛选[J]. 中国农业科学, 2017, 50(17): 3311-3322.
[11] 王阳,王文辉,贾晓辉,佟伟,王志华,杨晓龙. 梨不同品种果实冻藏品质性状分析与适宜品种筛选[J]. 中国农业科学, 2017, 50(17): 3400-3412.
[12] 尚杰,耿增超,王月玲,陈心想,赵军. 施用生物炭对 土微生物量碳、氮及酶活性的影响[J]. 中国农业科学, 2016, 49(6): 1142-1151.
[13] 张学林,徐 钧,安婷婷,侯小畔,李潮海. 不同氮肥水平下玉米根际土壤特性与产量的关系[J]. 中国农业科学, 2016, 49(14): 2687-2699.
[14] 屈洋,周瑜,王钊,王鹏科,高金锋,高小丽,冯佰利. 苦荞产区种质资源遗传多样性和遗传结构分析[J]. 中国农业科学, 2016, 49(11): 2049-2062.
[15] 刘亚秋,陈红艳,王瑞燕,常春艳,陈哲. 基于可见/近红外光谱的黄河口区土壤盐分及其主要离子的定量分析[J]. 中国农业科学, 2016, 49(10): 1925-1935.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!