[1]张晋恒,柏孝燚,伍金凤,等.基于高光谱和集成学习的人参果维生素C含量无损检测方法[J].江苏农业学报,2025,(09):1771-1780.[doi:doi:10.3969/j.issn.1000-4440.2025.09.012]
　ZHANG Jinheng,BAI Xiaoyi,WU Jinfeng,et al.A non-destructive detection method for vitamin C content in ginseng fruit based on hyperspectral data and ensemble learning[J].,2025,(09):1771-1780.[doi:doi:10.3969/j.issn.1000-4440.2025.09.012]
点击复制

基于高光谱和集成学习的人参果维生素C含量无损检测方法()

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年09期
页码:
1771-1780
栏目:
农业信息工程
出版日期:
2025-09-30

文章信息/Info

Title:
A non-destructive detection method for vitamin C content in ginseng fruit based on hyperspectral data and ensemble learning
作者:
张晋恒1; 柏孝燚1; 伍金凤2; 周兵1
(1.云南农业大学理学院,云南昆明650201；2.云南农业大学食品科学技术学院,云南昆明650201)
Author(s):
ZHANG Jinheng1; BAI Xiaoyi1; WU Jinfeng2; ZHOU Bing1
(1.College of Science, Yunnan Agricultural University, Kunming 650201, China;2.College of Food Science and Technology, Yunnan Agricultural University, Kunming 650201, China)
关键词:
人参果; 维生素C; 高光谱; 机器学习; 集成学习
Keywords:
ginseng fruit; vitamin C; hyperspectral; machine learning; ensemble learning
分类号:
TP181；TS255.7
DOI:
doi:10.3969/j.issn.1000-4440.2025.09.012
文献标志码:
A
摘要:
维生素C含量是评价人参果品质的重要指标,本研究通过获取人参果的高光谱数据,对人参果维生素C含量进行快速无损检测。为有效消除数据噪声的影响,采用移动平均平滑、多元散射校正、一阶导数、最小二乘平滑滤波进行光谱预处理,通过支持向量回归建模并对比预测效果,确定最优光谱预处理方法。针对高光谱数据特征降维的问题,采用竞争性自适应重加权算法、连续投影算法、轻量级梯度提升机算法提取与人参果维生素C含量高度相关的特征波长。将选定的特征波长结合支持向量回归、随机森林回归、多层感知机和Stacking方法进行建模并对比预测性能,确定最佳预测模型。结果表明,Stacking方法具有最佳预测性能,其验证集决定系数(R2)为0.917 2,均方根误差(RMSE)为15.053,相对预测误差（RPD）为3.595 3，该方法能够快速、准确地预测人参果维生素C含量,为人参果品质评价和分级分选提供技术支持。
Abstract:
Vitamin C content serves as a crucial index for evaluating the quality of ginseng fruits. In this study, rapid and non-destructive detection of vitamin C content in ginseng fruits was conducted by acquiring their hyperspectral data. To effectively eliminate the impact of data noise, spectral preprocessing methods such as moving average smoothing, multivariate scatter correction, first-order derivative, and least squares smoothing filter were employed. The optimal spectral preprocessing method was determined by comparing the predictive performance of models built using support vector regression. Addressing the issue of dimension reduction in hyperspectral data, competitive adaptive reweighted sampling, successive projections algorithm, and light gradient boosting machine algorithm were utilized to extract feature wavelengths highly correlated with the vitamin C content in ginseng fruits. The selected feature wavelengths were then combined with support vector regression, random forest regression, multi-layer perceptron, and Stacking methods for modeling and comparison to identify the best predictive model. The results indicated that the Stacking method exhibited the best prediction performance, with coefficient of determination (R2) of 0.917 2, root mean square error (RMSE) of 15.053, and relative percent deviation (RPD) of 3.595 3 for the validation set. This method enables rapid and accurate prediction of vitamin C content in ginseng fruits, providing technical support for the evaluation, grading, and sorting of ginseng fruit quality.

参考文献/References:

[1]RODRGUEZ-BURRUEZO A, KOLLMANNSBERGER H, PROHENS J, et al. Analysis of the volatile aroma constituents of parental and hybrid clones of pepino (Solanum muricatum)[J]. Journal of Agricultural and Food Chemistry,2004,52(18):5663-5669.
[2]王琰,张文刚,杨希娟,等. 人参果片干制方式及其品质特性研究[J]. 食品工业,2024,45(10):22-28.
[3]杜丽娟,黄兴龙,毕亚楠,等. 石林人参果的品质差异及综合性评价[J]. 食品安全质量检测学报,2023,14(23):107-114.
[4]张子琛,王玉英,张晚秋,等. 8个人参果品种(系)的果实品质评价[J]. 热带作物学报,2024,45(3):524-532.
[5]杨世鹏,蒋晓婷,许盼盼,等. 人参果营养成分、采后生理及贮藏保鲜方式研究进展[J]. 西北农业学报,2020,29(10):1447-1456.
[6]罗弦,王晓莉,罗娅,等.基于主成分分析的海藻肥对枇杷果实品质影响的综合评价[J].四川农业大学学报,2024,42(6):1212-1219.
[7]徐朝阳. 2,6-二氯酚靛酚滴定法与碘量法测定蔬菜水果中维生素C方法的准确度比较[J]. 食品安全导刊,2021(25):100-101.
[8]班兆军,高喧翔,马肄恒,等. 基于高光谱和深度学习的苹果品质无损检测方法[J]. 江苏农业学报,2024,40(8):1446-1454.
[9]LAN W J, JAILLAIS B, RENARD C M G C, et al. A method using near infrared hyperspectral imaging to highlight the internal quality of apple fruit slices[J]. Postharvest Biology and Technology,2021,175:111497.
[10]ZHU H Y, CHU B Q, FAN Y Y, et al. Hyperspectral imaging for predicting the internal quality of kiwifruits based on variable selection algorithms and chemometric models[J]. Scientific Reports,2017,7(1):7845.
[11]TIAN X, LI J B, WANG Q Y, et al. A multi-region combined model for non-destructive prediction of soluble solids content in apple,based on brightness grade segmentation of hyperspectral imaging[J]. Biosystems Engineering,2019,183:110-120.
[12]PAN L Q, ZHANG Q, ZHANG W, et al. Detection of cold injury in peaches by hyperspectral reflectance imaging and artificial neural network[J]. Food Chemistry,2016,192:134-141.
[13]HE H J, ZHANG C, BIAN X H, et al. Improved prediction of vitamin C and reducing sugar content in sweetpotatoes using hyperspectral imaging and LARS-enhanced LASSO variable selection[J]. Journal of Food Composition and Analysis,2024,132:106350.
[14]FATCHURRAHMAN D, NOSRATI M, AMODIO M L, et al. Comparison performance of visible-NIR and near-infrared hyperspectral imaging for prediction of nutritional quality of goji berry (Lycium barbarum L. )[J]. Foods,2021,10(7):1676.
[15]张伏,曹炜桦,崔夏华,等. 基于SG-CARS-IBP的圣女果可溶性固形物可见/近红外光谱无损检测[J]. 光谱学与光谱分析,2023,43(3):737-743.
[16]袁旭林. 基于高光谱成像技术的苹果糖度无损检测系统研究[D]. 济南:山东大学,2021.
[17]高梦蕾. 三种赏食两用植物的无土栽培技术研究[D]. 哈尔滨:东北农业大学,2018.
[18]郭林鸽,殷勇,于慧春,等. 基于Fisher判别分析可分性信息融合的马铃薯VC含量高光谱检测方法[J]. 食品科学,2024,45(7):164-171.
[19]MISHRA P, KARAMI A, NORDON A, et al. Automatic de-noising of close-range hyperspectral images with a wavelength-specific shearlet-based image noise reduction method[J]. Sensors and Actuators B:Chemical,2019,281:1034-1044.
[20]李奇辰,李民赞,杨玮,等. 基于拉曼光谱的水溶性磷定量分析[J]. 光谱学与光谱分析,2023,43(12):3871-3876.
[21]GUO W, LI X X, XIE T H. Method and system for nondestructive detection of freshness in Penaeus vannamei based on hyperspectral technology[J]. Aquaculture,2021,538:736512.
[22]DPPER V, ROCHA A D, BERGER K, et al. Estimating soil moisture content under grassland with hyperspectral data using radiative transfer modelling and machine learning[J]. International Journal of Applied Earth Observation and Geoinformation,2022,110:102817.
[23]MA L, ZHANG Y, ZHANG Y Y, et al. Rapid nondestructive detection of chlorophyll content in muskmelon leaves under different light quality treatments[J]. Agronomy,2022,12(12):3223.
[24]LI H D, LIANG Y Z, XU Q S, et al. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration[J]. Analytica Chimica Acta,2009,648(1):77-84.
[25]ARAUJO M C U, SALDANHA T C B, GALVAO R K H, et al. The successive projections algorithm for variable selection in spectroscopic multicomponent analysis[J]. Chemometrics and Intelligent Laboratory Systems,2001,57(2):65-73.
[26]吴继忠,时艺丹,黄慧,等. 基于改进堆叠自编码器结合LightGBM的近红外光谱回归算法研究[J]. 分析测试学报,2023,42(9):1112-1118.
[27]DRUCKER H, BURGES C J C, KAUFMAN L, et al. Support vector regression machines[C]//MOZER M C, JORDAN M, PETSCHE T. Advances in Neural Information Processing Systems 9. Cambridge:MIT Press,1996:155-161.
[28]张驰,郭媛,黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用,2021,57(11):57-69.
[29]BREIMAN L. Random forests[J]. Machine Learning,2001,45(1):5-32.
[30]VISCARRA ROSSEL R A, MCGLYNN R N, MCBRATNEY A B. Determining the composition of mineral-organic mixes using UV-vis-NIR diffuse reflectance spectroscopy[J]. Geoderma,2006,137(1/2):70-82.
[31]刘子涵,李明,赵峙尧,等. 基于高光谱成像技术和机器学习的猕猴桃果实可溶性固形物含量预测[J]. 果树学报,2024,41(12):2606-2620.
[32]MISHRA P, WOLTERING E, BROUWER B, et al. Improving moisture and soluble solids content prediction in pear fruit using near-infrared spectroscopy with variable selection and model updating approach[J]. Postharvest Biology and Technology,2021,171:111348.
[33]LI X L, WEI Y Z, XU J, et al. SSC and pH for sweet assessment and maturity classification of harvested cherry fruit based on NIR hyperspectral imaging technology[J]. Postharvest Biology and Technology,2018,143:112-118.
[34]郑艺蕾. 基于高光谱和太赫兹光谱的甘薯品质检测方法研究[D]. 南昌:华东交通大学,2020.
[35]王世芳,韩平,崔广禄,等. SPXY算法的西瓜可溶性固形物近红外光谱检测[J]. 光谱学与光谱分析,2019,39(3):738.
[36]宋子怡,常庆瑞,郑智康,等. 基于高光谱和连续投影算法的猕猴桃叶片氮平衡指数的估测[J]. 江苏农业学报,2024,40(7):1260-1267.
[37]陶惠林,冯海宽,杨贵军,等. 基于无人机数码影像和高光谱数据的冬小麦产量估算对比[J]. 农业工程学报,2019,35(23):111-118.
[38]段丹丹,刘仲华,赵春江,等. 基于特征光谱参数的叶片和冠层尺度茶多酚含量估算[J]. 光谱学与光谱分析,2024,44(3):814-820.

备注/Memo

备注/Memo:
收稿日期：2025-01-25基金项目：云南省重大科技专项(202302AE09002003)作者简介：张晋恒(1979-),男,云南建水人,硕士,讲师,主要从事高光谱技术在农业食品中的应用研究。(E-mail)zhang_zhw@163.com通讯作者：周兵,(E-mail)bingzhoukm@126.com
更新日期/Last Update: 2025-10-27