Building hyperspectral model of oleic acid content in rapeseed of vegetable and oil type Brassica napus
-
摘要: 【目的】基于高光谱特征初步判别油菜摘薹情况,为实现高光谱反演籽粒油酸含量提供理论指导。【方法】使用FieldSpec 3地物光谱仪采集油菜盛花期叶片光谱数据,采用Agilent GC-MS 7980B气相色谱仪分析摘薹和未摘薹处理的籽粒油酸含量,比较2组处理的平均原始光谱反射率特征,及其油菜叶片原始及一阶微分光谱反射率与籽粒油酸含量相关性,在此基础上构建基于原始光谱特征波长的支持向量机(SVM)判别模型、基于光谱参数的油酸含量二项式模型、基于一阶微分光谱特征波长的油酸含量多元线性逐步回归(MLSR)及偏最小二乘回归(PLSR)预测模型,并利用独立样本T检验对模型精度进行验证。【结果】发现未摘薹及摘薹处理的平均原始光谱反射率曲线在760~1080nm波段存在一定差异。未摘薹及摘薹处理的原始光谱反射率与籽粒油酸含量相关性曲线存在一定差异,未摘薹处理的原始光谱反射率在484~956和1001~1146 nm波段与籽粒油酸含量呈正相关,摘薹处理的原始光谱反射率在1882~2111和2324~2499 nm波段与油菜籽粒油酸含量呈正相关,说明摘薹会影响油菜光谱反射率与籽粒油酸含量的相关性表现。选取位于760~1080 nm波段4个拐点波长(760、920、970和1080 nm)的原始光谱反射率作为自变量,用以构建SVM判别模型,经过多次随机取样比较构建所有SVM判别模型,发现最佳判别模型的训练集样本总体精度为86.1%,验证集样本总体精度为77.8%,说明利用高光谱技术判别油菜是否摘薹具有一定的可行性。光谱参数模型中RVI模型对未摘薹处理油菜籽粒油酸含量的反演效果最佳,且该模型与未摘薹处理籽粒油酸含量的相关系数(-0.705)最高。比较全部油菜籽粒油酸含量预测模型类型,PLSR模型对未摘薹处理籽粒油酸含量预测精度最高,其训练集R2=0.590、RMSE=0.610,MLSR模型对摘薹处理籽粒油酸含量预测精度最高,其训练集R2=0.773、RMSE=0.874。利用独立样本T检验对二者模型测试集样本进行验证,未摘薹样本P=0.839,摘薹样本P=0.858,二者样本实测值与预测值均无显著差异(P>0.05),模型合理,说明利用高光谱技术对油菜籽粒油酸含量进行预测可行。【建议】引入随机森林等机器学习算法,更好地选取特征波长(显著相关波长或全波段等),提高光谱数据对油菜籽粒油酸含量的预测能力。后期的试验应侧重于多品种油菜籽粒油酸含量估测研究,探索高光谱技术估测油菜籽粒油酸含量是否具备普遍的可行性。利用高光谱技术反演其他油菜籽粒品质指标,为高光谱遥感监测油菜品质提供理论依据。Abstract: 【Objective】Based on the hyperspectral characteristics of Brassica napus, a preliminary discrimination of rape moss picking was conducted, and provided theoretical guidance for the realization of hyperspectral inversion of rapeseed oleic acid content.【Method】The spectral data of B. napus leaves were collected by FieldSpec 3, and the oleic acid content of rapeseed with mossed treatment and unmossed treatment was analyzed by Agilent GC-MS 7980 B gas chromatograph. The average original spectral reflectance characteristics of the two treatments were compared. The correlation between the original and first-order differential spectral reflectance and the oleic acid content of rapeseed with and without moss picking was analyzed. On the basis of the original spectrum characteristic wavelength based support vector machine(SVM) discriminant model, based on spectral parameters of the oleic acid content in binomial model, based on the firstorder spectrum characteristic wavelengths of oleic acid content(MLSR) with multiple linear stepwise regression and partial least-squares regression(PLSR) prediction model, independent samples T test was used to verify this model precision.【Result】It was found that the average original spectral reflectance curves of unpicked and moss-picked rapeseed were different in the band range of 760-1080 nm. There were some differences between the original spectral reflectance and the oleic acid content of rapeseed between unplucked and plucked rapeseed. The original spectral reflectance of unplucked rapeseed was 484-956 nm and 1001-1146 nm, and the original spectral reflectance of plucked rapeseed was 1882-2111 nm and 2324-2499 nm, indicating that the correlation between spectral reflectance and oleic acid content of rapeseed was affected by moss picking. The original spectral reflectance at four inflection points(760, 920, 970 and 1080 nm) in the 760-1080 nm band was selected as the independent variable to construct the SVM discriminant model. After multiple random sampling and comparison of all SVM discriminant models, it was found that the overall accuracy of the training set sample of the best discriminant model was 86.1%. The overall accuracy of the validation set was 77.8%, which indicated that it was feasible to use hyperspectral technology to determine whether rapeseed moss picking. Among the spectral parameter models, RVI model had the best inversion effect on the content of rapeseed oleic acid in unmossed rapeseed, and the correlation coefficient between RVI and the content of rapeseed oleic acid in unmossed rapeseed was-0.705. Comparing the prediction models of oleic acid content in rapeseed of all types, PLSR model had the highest prediction accuracy of oleic acid content in rapeseed of unmossed rapeseed, with its training set R2=0.590 and RMSE=0.610;MLSR model had the highest prediction accuracy of oleic acid content in rapeseed of mossed rapeseed, with its training set R2=0.773 and RMSE=0.874. The independent sample T test was used to verify the two model test sets of samples. P=0.839 for the unmossed sample and P=0.858 for the mossed sample, and there was no significant difference between the measured and predicted values of the two samples(P>0.05). The model was reasonable, indicating that it was feasible to predict the oleic acid content of rapeseed using hyperspectral technology.【Suggestion】Introduce random forest and other machine learning algorithms to better select characteristic wavelengths(significantly correlated wavelengths or full bands, etc.) and improve the prediction ability of spectral data for rapeseed oleic acid content. Later experiments should focus on the estimation of rapeseed oleic acid content in multiple varieties of B. napus, and explore whether hyperspectral technology is universally feasible to estimate rapeseed oleic acid content. Other rapeseed quality indexes were retrieved by hyperspectral technique, which provides the basis for monitoring rapeseed quality by hyperspectral remote sensing.
-
-
常涛, 王悦, 程潜, 张振乾, 官春云, 王国槐.2018.高光谱反射法早期快速筛选高含油量油菜种质资源[J]. 激光生物学报, 27(6):510-515.doi:10.3969/j.issn.1007-7146.2018.06.004.[Chang T, Wang Y, Cheng Q, Zhang Z Q, Guan C Y, Wang G H.2018.Early screening of rapeseed germplasm for high oil content by hyperspectral reflectance method[J]. Acta Laser Biology Sinica, 27(6):510-515.] 谌俊旭.2019.不同施氮水平下带状套作玉米冠层光谱特征及产量品质模型研究[D].成都:四川农业大学.[Chen J X.2019.Canopy hyperspectral characteristics and model of yield and quality of maize under different nitrogen fertilizer levels in intercropping system[D].Chengdu:Sichuan Agricultural University.] 程潜, 李斌, 张振乾, 官春云, 王悦, 陈浩, 方希林, 杨鑫.2018.不同硼用量下油菜冠层反射光谱与叶绿素含量间定量关系[J]. 西南农业学报, 31(10):2127-2134.doi:10.16213/j.cnki.scjas.2018.10.023.[Cheng Q, Li B, Zhang Z Q, Guan C Y, Wang Y, Chen H, Fang X L, Yang X.2018.Quantitative relationship between canopy reflectance spectra and chlorophyll content of rape under different boron content[J]. Southwest China Journal of Agricultural Sciences, 31(10):2127-2134.] 崔小涛, 常庆瑞, 屈春燕, 史博太, 蒋丹垚, 夏利恒, 王玉娜.2020.基于高光谱和MLSR-GA-BP神经网络模型油菜叶片SPAD值遥感估算[J]. 东北农业大学学报, 51(8):74-84.doi:10.19720/j.cnki.issn.1005-9369.2020.08.0010.[Cui X T, Chang Q R, Qu C Y, Shi B T, Jiang D Y, Xia L H, Wang Y N.2020.Remote sensing estimation of SPADvalue for rape leaf based on hyperspectral and MLSR-GA-BP neural network model[J]. Journal of Northeast Agricultural University, 51(8):74-84.] 高开秀, 高雯晗, 明金, 李岚涛, 汪善勤, 鲁剑巍.2019.无人机载多光谱遥感监测冬油菜氮素营养研究[J]. 中国油料作物学报, 41(2):232-242.doi:10.7505/j.issn.1007-9084.2019.02.011.[Gao K X, Gao W H, Ming J, Li L T, Wang S Q, Lu J W.2019.Monitoring of nitrogen nutrition in winter rapeseed using UAV-borne multispectral data[J]. Chinese Journal of Oil Crop Sciences, 41(2):232-242.] 龚龑, 肖洁, 候金雨, 段博.2017.基于无人机遥感混合光谱分析的油菜估产模型[J]. 测绘地理信息, 42(6):40-45.doi:10.14188/j.2095-6045.2016350.[Gong Y, Xiao J, Hou J Y, Duan B.2017.Rape yields estimation research based on spectral analysis for UAV image[J]. Journal of Geomatics, 42(6):40-45.] 郭滢, 崔看, 覃磊, 张小波, 童建华, 刘忠松, 夏石头.2019.芥甘杂交黄籽油菜籽粒中植物激素的变化及其对脂肪酸成分与油脂含量的影响[J]. 激光生物学报, 28(6):548-555.doi:10.3969/j.issn.1007-7146.2019.06.011.[Guo Y, Cui K, Qin L, Zhang X B, Tong J H, Liu Z S, Xia S T.2019.Dynamic change of phytohormones in seeds of yellow-seeded canola progeny from hybridization between Brassica juncea and Brassica napus and its effects on fatty acid composition and oil content[J]. Acta Laser Biology Sinica, 28(6):548-555.] 栗方亮, 孔庆波, 张青.2020.光谱分析技术在作物氮素诊断上的应用研究进展[J]. 江西农业学报, 32(9):6-12.doi:10.19386/j.cnki.jxnyxb.2020.09.02.[Li F L, Kong QB, Zhang Q.2020.Research progress in application of spectral analysis technology in crop nitrogen diagnosis[J]. Acta Agriculturae Jiangxi, 32(9):6-12.] 林宝刚, 任韵, 柳寒, 何伟民, 张冬青, 吴德志, 余华胜, 华水金.2020.不同生态区对油菜薹营养品质的影响及摘薹后产量表现[J]. 核农学报, 34(9):2071-2079.doi:10.11869/j.issn.100-8551.2020.09.2071.[Lin B G, Ren Y, Liu H, He W M, Zhang D Q, Wu D Z, Yu H S, Hua S J.2020.Effects of ecological sites on the nutritional quality of young stem of rapeseed and yield performance after post-harvesting[J]. Journal of Nuclear Agricultural Sciences, 34(9):2071-2079.] 刘红梅, 肖正午, 申涛, 蒋鹏, 单双吕, 邹应斌.2019.稻米直链淀粉含量近红外检测模型的建立[J]. 湖南农业大学学报(自然科学版), 45(2):189-193.doi:10.13331/j.cnki.jhau.2019.02.014.[Liu H M, Xiao Z W, Shen T, Jiang P, Shan S L, Zou Y B.2019.Establishment of near infrared analysis model for amylose content of rice powder[J]. Journal of Hunan Agricultural University(Natural Science), 45(2):189-193.] 卢俊玮, 田容才, 花宇辉, 官春云.2020.光谱技术在甘蓝型油菜品质检测中的研究进展[J]. 激光生物学报, 29(4):289-294.doi:10.3969/j.issn.1007-7146.2020.04.001.[Lu J W, Tian R C, Hua Y H, Guan C Y.2020.Research progress of spectral technology in detection of quality in oilseed rape(Brassica napus L.)[J]. Acta Laser Biology Sinica, 29(4):289-294.] 施光耀, 王若伦, 桑玉强, 张劲松, 孟平.2020.基于高光谱的核桃-大豆复合系统大豆叶绿素估算研究[J]. 河南农业大学学报, 54(5):762-769.doi:10.16445/j.cnki.1000-2340.2020.05.005.[Shi G Y, Wang R L, Sang Y Q, Zhang J S, Meng P.2020.Estimation of soybean chlorophyll in walnut-soybean hybrid system based on hyperspectral[J]. Journal of Henan Agricultural University, 54(5):762-769.] 孙玉婷, 杨红云, 孙爱珍, 梅芳, 易文龙.2020.水稻叶片SPAD值的高光谱估算模型[J]. 南方农业学报, 51(5):1062-1069.doi:10.3969/j.issn.2095-1191.2020.05.010.[Sun YT, Yang H Y, Sun A Z, Mei F, Yi W L.2020.Hyperspectral estimation model of rice leaf SPAD value[J]. Journal of Southern Agriculture, 51(5):1062-1069.] 田容才, 高志强, 卢俊玮.2020.基于冠层光谱的早籼稻籽粒粗蛋白含量估测[J]. 作物杂志, (4):188-194.doi:10.16035/j.issn.1001-7283.2020.04.027.[Tian R C, Gao Z Q, Lu JW.2020.Estimation of crude protein content in early indica rice grains based on canopy spectrum[J]. Crops Journal, (4):188-194.] 汪波, 罗元庆, 胡英哲, 黄俊生, 闾晓平, 甘丽, 李泽圆, 阿旺德吉, 蒯婕, 傅廷栋, 周广生.2019."一菜两用"油菜菜叶采摘时间和留叶数量效益研究[J]. 长江蔬菜, (14):33-35.doi:10.3865/j.issn.1001-3547.2019.14.014.[Wang B, Luo Y Q, Hu Y Z, Huang J S, LüX P, Gan L, Li Z Y, AW D J, Kuai J, Fu T D, Zhou G S.2019.Study on the efficiency of "one vegetable with two applications" in the picking time and the quantity of leaves left in rape[J]. Journal of Changjiang Vegetables, (14):33-35.] 邢璐琪, 张桂莲, 张浪, 易扬, 林勇, 江子尧, 郑谐维.2020.高光谱技术在植被特征监测中的应用[J]. 园林, (11):8-14.doi:10.12193/j.laing.2020.11.0008.002.[Xing L Q, Zhang G L, Zhang L, Yi Y, Lin Y, Jiang Z Y, Zheng XW.2020.Application of hyperspectral technique in monitoring vegetation characteristics[J]. Landscape Architecture, (11):8-14.] 徐扬帆, 张秋平, 徐明水, 张振乾.2020.高含油高油酸油菜新品种‘帆鸣1号’选育及品质分析[J]. 分子植物育种, 18(15):5184-5190.doi:10.13271/j.mpb.018.005184.[Xu Y F, Zhang Q P, Xu M S, Zhang Z Q.2020.The breeding and quality analysis of a new rape variety ‘Fanming1’ with high oil content and oleic acid content[J]. Molecular Plant Breeding, 18(15):5184-5190.] 尹凡, 刘燕, 董晶晶, 卢长明, 徐海根.2013.基于冠层实测光谱的油菜品种识别[J]. 湖北农业科学, 52(23):5866-5870.doi:10.14088/j.cnki.issn0439-8114.2013.23.060.[Yin F, Liu Y, Dong J J, Lu C M, Xu H G.2013.Identification of rapeseed varieties based on canopy measurement spectrum[J]. Hubei Province Agricultural Science, 52(23):5866-5870.] 赵赞忠.2018.油菜冠层特征光谱参数及其叶面积指数估测研究[D].长沙:湖南农业大学.[Zhao Z Z.2018.Estimation of canola canopy characteristic spectral parameters and leaf area index[D].Changsha:Hunan Agricultural University.] Das B, Manohara K K, Mahajan G R, Sahoo R N.2020.Spectroscopy based novel spectral indices, PCA-and PLSR-coupled machine learning models for salinity stress phenotyping of rice[J]. Spectrochimica Acta Part A:Mole-cular and Biomolecular Spectroscopy, 229:117983.doi: 10.1016/j.saa.2019.117983.
Duan B, Fang S H, Zhu R S, Wu X T, Wang S Q, Gong Y, Peng Y.2019.Remote estimation of rice yield with unmanned aerial vehicle(UAV) data and spectral mixture analysis[J]. Frontiers in Plant Science, 10:204.doi: 10.3389/fpls.2019.00204.
Li L T, Ren T, Ma Y, Wei Q Q, Wang S H, Li X K, Cong RH, Liu S S, Lu J W.2016.Evaluating chlorophyll density in winter oilseed rape(Brassica napus L.) using canopy hyperspectral red-edge parameters[J]. Computers and Electronics in Agriculture, 126:21-31.doi: 10.1016/j.compag.2016.05.008.
Shi Y F, Lou Y S, Zhang Z, Ma L, Ojara M.2020.Estimation of methane emissions based on crop yield and remote sensing data in a paddy field[J]. Greenhouse Gases:Science and Technology, 10(1):196-207.doi: 10.1002/ghg.1946.
计量
- 文章访问数: 62
- HTML全文浏览量: 3
- PDF下载量: 1