南瓜干燥过程中β-胡萝卜素的多源融合预测

doi:10.3969/j.issn.1004-1524.20221004

浙江农业学报 ›› 2023, Vol. 35 ›› Issue (8): 1876-1887.DOI: 10.3969/j.issn.1004-1524.20221004

南瓜干燥过程中β-胡萝卜素的多源融合预测

宁文楷¹^,²(), 李静¹^,², 沈晓东¹^,², 吴鑫¹, 李臻锋¹^,²^,^*()

1.江南大学机械工程学院,江苏无锡 214122
2.江苏省食品先进制造装备技术重点实验室,江苏无锡 214122

收稿日期:2022-07-06 出版日期:2023-08-25 发布日期:2023-08-29
作者简介:宁文楷(1997—),男,吉林辽源人,硕士研究生,研究方向为食品无损检测。E-mail:17766480325@163.com
通讯作者: *李臻锋,E-mail:1736691239@qq.com
基金资助:
国家自然科学基金(51508229);江苏省普通高校自然科学研究计划项目(KYCX19_1862)

Prediction of multi-source fusion of β-carotene during pumpkin drying

NING Wenkai¹^,²(), LI Jing¹^,², SHEN Xiaodong¹^,², WU Xin¹, LI Zhenfeng¹^,²^,^*()

1. School of Mechanical Engineering, Jiangnan University, Wuxi 214122, Jiangsu, China
2. Jiangsu Key Laboratory of Advanced Food Manufacturing Equipment and Technology, Wuxi 214122, Jiangsu, China

Received:2022-07-06 Online:2023-08-25 Published:2023-08-29

摘要/Abstract

摘要：

β-胡萝卜素是南瓜品质的重要指标,传统检测方法周期长且过程复杂。为实现南瓜微波干燥过程中的β-胡萝卜素含量预测,本研究搭建了包含含水率检测单元、机器视觉单元和电子鼻单元的微波干燥系统,在线检测60、70、80 ℃不同温度下南瓜干燥过程中的含水率、外观形态(表面皱缩率、色差)和气味特征,同时检测β-胡萝卜素含量。结果表明,南瓜干燥过程中β-胡萝卜素含量与外观形态和气味特征间呈现显著相关。通过建立单源(机器视觉或电子鼻)和多源融合(机器视觉融合电子鼻)极限学习机模型,对南瓜干燥过程中β-胡萝卜素含量进行预测,预测结果表明,单源模型中机器视觉模型比电子鼻模型有更好的预测效果,且多源融合模型预测精度最高, $R p 2$ >0.97。基于多源融合的极限学习机模型可以有效预测南瓜微波干燥过程中β-胡萝卜素含量,在南瓜自动化加工及品质快速监测方面具有良好的应用前景。

关键词: 南瓜, 机器视觉, 电子鼻, β-胡萝卜素, 预测

Abstract:

β-carotene is an important index of pumpkin quality. The traditional method has a long period and a complicated process. To achieve the prediction of β-carotene content in pumpkin in the process of microwave drying, this study built a microwave drying system containing moisture content detection unit, machine vision and electronic nose units, the moisture content and appearance shape (surface shrinkage rate, and color difference) and odor characteristics of pumpkin under different temperatures of 60, 70, 80 ℃ in the drying process were on-line detected, meanwhile, the content of β-carotene was detected. The results showed that there was a significant correlation between the content of β-carotene and the appearance and flavor characteristics of pumpkin during drying. By establishing a monophyletic (machine vision or electronic nose) and multi-source fusion extreme learning machine (machine vision fusion electronic nose) model, the β-carotene content of pumpkin in the process of drying was predicted, and the results showed that the monophyletic model in machine vision had better prediction effect than electronic nose model, and the prediction accuracy of multi-source fusion model was the highest, $R p 2$ >0.97. The extreme learning machine model based on multi-source fusion can effectively predict the content of β-carotene in microwave drying of pumpkin, which has a good application prospect in automatic processing and quality monitoring of pumpkin.

Key words: pumpkin, machine vision, electronic nose, β-carotene, prediction

中图分类号:

TS207.3

宁文楷, 李静, 沈晓东, 吴鑫, 李臻锋. 南瓜干燥过程中β-胡萝卜素的多源融合预测[J]. 浙江农业学报, 2023, 35(8): 1876-1887.

NING Wenkai, LI Jing, SHEN Xiaodong, WU Xin, LI Zhenfeng. Prediction of multi-source fusion of β-carotene during pumpkin drying[J]. Acta Agriculturae Zhejiangensis, 2023, 35(8): 1876-1887.

图/表 17

图1 试验系统示意图

Fig.1 Schematic diagram of the test system

表1 PEN3电子鼻传感器及其性能描述

Table 1 PEN3 electronic nose sensor and its performance description

传感器 Sensor	传感器型号 Sensor type	敏感气体类型 Sensitive gas type
1	W1C	对芳香物族化合物灵敏Sensitive to aromatic compounds
2	W5S	对含氮氧类物质灵敏Sensitive to nitrogen and oxygen containing substances
3	W3C	对氨等芳香物族化合物灵敏Sensitive to aromatic compounds such as ammonia
4	W6S	对氢气灵敏Sensitive to hydrogen gas
5	W5C	对烷烃、芳香族化合物灵敏Sensitive to alkanes and aromatic compounds
6	W1S	对甲基类物质灵敏Sensitive to methyl substances
7	W1W	对有机硫化物灵敏Sensitive to organic sulfides
8	W2S	对含有芳香族化合物的醇类灵敏Sensitive to alcohols containing aromatic compounds
9	W2W	对芳香成分灵敏Sensitive to aromatic components
10	W3S	对高浓度烷烃-脂肪族物质灵敏Sensitive to high concentrations of alkanes and aliphatic substances

图2 不同干燥温度对南瓜干燥特性影响

Fig.2 Effects of different drying temperatures on drying characteristics of pumpkin

图3 不同干燥温度对南瓜β-胡萝卜素含量影响

Fig.3 Effects of different drying temperatures on β-carotene content of pumpkin

图4 不同干燥温度下南瓜外观形态采集图像

Fig.4 Collected images of pumpkin appearance at different drying temperatures

图5 不同干燥温度对南瓜表面皱缩的影响 a,纵截面;b,横截面。

Fig.5 Effect of different drying temperatures on surface shrinkage of pumpkin a, Longitudinal section; b, Cross section.

图6 不同干燥温度对南瓜色差影响

Fig.6 Influence of different drying temperatures on pumpkin color difference

图7 南瓜鲜样电子鼻传感器雷达图

Fig.7 Pumpkin fresh sample electronic nose sensor radar

图8 南瓜干燥过程中各传感器响应响应值变化

Fig.8 Changes in response values of each sensor during pumpkin drying

图9 不同干燥温度对电子鼻传感器响应值影响 a, 传感器2;b,传感器6;c,传感器7;d,传感器9。

Fig.9 Influence of different drying temperatures on response value of electronic nose sensor a, Sensor 2; b, Sensor 6; c, Sensor 7; d, Sensor 9.

图10 β-胡萝卜素动力学拟合图 a,一级动力学;b, 二级动力学;c,三级动力学。

Fig.10 Kinetic fitting of β-carotene a; First-order dynamics; b, Second-order dynamics; c, Third-order dynamics.

表2 β-胡萝卜素动力学相关系数

Table 2 Kinetic correlation coefficient of β-carotene

温度 Temperature/ ℃	一级动力学First-order dynamics			二级动力学Second-order dynamics			三级动力学Third-order dynamics
温度 Temperature/ ℃	k/min^-1	R²	RMSE	k/(mg· kg^-1)^-1·min^-1	R²	RMSE	k/(mg· kg^-1)^-2·min^-1	R²	RMSE
60	0.005 83	0.858 74	0.020 29	2.55×10^-4	0.901 23	2.39×10^-5	1.189×10^-5	0.940 91	2.975×10^-8
70	0.006 69	0.891 46	0.010 33	2.48×10^-4	0.917 25	1.049×10^-5	9.533×10^-6	0.932 85	1.240×10^-8
80	0.011 73	0.957 69	0.006 27	5.15×10^-4	0.985 39	4.060×10^-6	2.409×10^-6	0.980 99	1.160×10^-8

表3 β-胡萝卜素含量与特征间相关性

Table 3 Correlation between β-carotene content and characteristics

温度 Temperature/ ℃	品质指标 Quality index	横截面皱缩率 Cross section shrinkage rate	纵截面皱缩率 Longitudinal section shrinkage rate	色差 Chromatism	传感器2 Sensor 2	传感器6 Sensor 6	传感器7 Sensor 7	传感器9 Sensor 9
60	β-胡萝卜素含量β-Carotene content	-0.948	-0.959	-0.904	0.915	0.923	0.870	0.894
70	β-胡萝卜素含量β-Carotene content	-0.933	-0.936	-0.979	0.742	0.977	0.952	0.769
80	β-胡萝卜素含量β-Carotene content	-0.973	-0.973	-0.984	0.933	0.919	0.888	0.938

图11 60 ℃下3种模型对β-胡萝卜素含量的预测散点图

Fig.11 Scatter plot of β-carotene content predicted by the three models at 60 ℃

表4 不同模型对β-胡萝卜素含量预测结果

Table 4 Prediction results of β-carotene content in different models

模型 Model	60 ℃		70 ℃		80 ℃
模型 Model	预测集 $R p 2$	预测集根均方差 RMSEP	预测集 $R p 2$	预测集根均方差 RMSEP	预测集 $R p 2$	预测集根均方差 RMSEP
机器视觉Machine vision	0.966 7	0.981 2	0.978 5	0.962 3	0.961 2	0.956 8
电子鼻Electronic nose	0.938 1	0.904 3	0.915 4	1.395 2	0.905 0	1.016 6
多源融合Multi-source fusion	0.981 3	0.942 3	0.985 2	0.854 7	0.979 5	0.924 2

表4 不同模型对β-胡萝卜素含量预测结果

Table 4 Prediction results of β-carotene content in different models

模型 Model	60 ℃		70 ℃		80 ℃
模型 Model	预测集 $R p 2$	预测集根均方差 RMSEP	预测集 $R p 2$	预测集根均方差 RMSEP	预测集 $R p 2$	预测集根均方差 RMSEP
机器视觉Machine vision	0.966 7	0.981 2	0.978 5	0.962 3	0.961 2	0.956 8
电子鼻Electronic nose	0.938 1	0.904 3	0.915 4	1.395 2	0.905 0	1.016 6
多源融合Multi-source fusion	0.981 3	0.942 3	0.985 2	0.854 7	0.979 5	0.924 2

图12 70 ℃下3种模型对β-胡萝卜素含量的预测散点图

Fig.12 Scatter plot of β-carotene content predicted by the three models at 70 ℃

图13 80 ℃下3种模型对β-胡萝卜素含量的预测散点图

Fig.13 Scatter plot of β-carotene content predicted by the three models at 80 ℃

参考文献 25

[1]	白竣文, 周存山, 蔡健荣, 等. 南瓜片真空脉动干燥特性及含水率预测[J]. 农业工程学报, 2017, 33(17): 290-297.
	BAI J W, ZHOU C S, CAI J R, et al. Vacuum pulse drying characteristics and moisture content prediction of pumpkin slices[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(17): 290-297. (in Chinese with English abstract)
[2]	LI J, LI Z F, RAGHAVAN G S V, et al. Fuzzy logic control of relative humidity in microwave drying of hawthorn[J]. Journal of Food Engineering, 2021, 310: 110706.
[3]	袁越锦, 洪晨, 徐英英, 等. 组合干燥方式下胡萝卜真空脉动蒸汽烫漂工艺优化[J]. 浙江农业学报, 2022, 34(1):163-172.
	YUAN Y J, HONG C, XU Y Y, et al. Optimization of carrot vacuum pulsating steam blanching process under combined drying method[J]. Acta Agriculturae Zhejiangensis, 2022, 34(1):163-172. (in Chinese with English abstract)
[4]	DE CARVALHO L M J, GOMES P B, DE OLIVEIRA GODOY R L, et al. Total carotenoid content, α-carotene and β-carotene, of Landrace pumpkins (Cucurbita moschata Duch): a preliminary study[J]. Food Research International, 2012, 47(2): 337-340.
[5]	LI C Y, CAO Q X, GUO F. A method for color classification of fruits based on machine vision[J]. WSEAS Transactions on Systems, 2009, 8(2): 312-321.
[6]	ZHU H K, YE Y, HE H F, et al. Evaluation of green tea sensory quality via process characteristics and image information[J]. Food and Bioproducts Processing, 2017, 102: 116-122.
[7]	CHEN C, LIU B Y, SONG F H, et al. An adaptive fuzzy logic control of green tea fixation process based on image processing technology[J]. Biosystems Engineering, 2022, 215: 1-20.
[8]	YADAV S P, IBARAKI Y, GUPTA S D. Estimation of the chlorophyll content of micropropagated potato plants using RGB based image analysis[J]. Plant Cell, Tissue and Organ Culture, 2010, 100(2): 183-188.
[9]	GUPTA S D, IBARAKI Y, PATTANAYAK A K. Development of a digital image analysis method for real-time estimation of chlorophyll content in micropropagated potato plants[J]. Plant Biotechnology Reports, 2013, 7(1): 91-97.
[10]	SIMKIN A J, UNDERWOOD B A, AULDRIDGE M, et al. Circadian regulation of the PhCCD1 carotenoid cleavage dioxygenase controls emission of β-ionone, a fragrance volatile of Petunia flowers[J]. Plant Physiology, 2004, 136(3): 3504-3514.
[11]	徐海霞. 基于机器视觉和电子鼻技术的菠菜新鲜度无损检测研究[D]. 镇江: 江苏大学, 2016.
	XU H X. Nondestructive detection of spinach freshness based on machine vision and electronic nose technology[D]. Zhenjiang: Jiangsu University, 2016. (in Chinese with English abstract)
[12]	王钊, 李长滨, 王源, 等. 多种蔬菜中β-胡萝卜素含量测定及油脂烹制作用[J]. 中国调味品, 2021, 46(2): 61-64.
	WANG Z, LI C B, WANG Y, et al. Determination of β-carotene content in various vegetables and its oil cooking effect[J]. China Condiment, 2021, 46(2): 61-64. (in Chinese)
[13]	SHAHABI M, RAFIEE S, MOHTASEBI S S, et al. Image analysis and green tea color change kinetics during thin-layer drying[J]. Food Science and Technology International, 2014, 20(6): 465-476.
[14]	LI X Y, LIU Y H, GAO Z J, et al. Computer vision online measurement of shiitake mushroom (Lentinus edodes) surface wrinkling and shrinkage during hot air drying with humidity control[J]. Journal of Food Engineering, 2021, 292: 110253.
[15]	DAI J W, RAO J Q, WANG D, et al. Process-based drying temperature and humidity integration control enhances drying kinetics of apricot halves[J]. Drying Technology, 2015, 33(3): 365-376.
[16]	RAGHAVAN G S V, LI Z F, WANG N, et al. Control of microwave drying process through aroma monitoring[J]. Drying Technology, 2010, 28(5): 591-599.
[17]	SUVARNAKUTA P, DEVAHASTIN S, MUJUMDAR A S. Drying kinetics and β-carotene degradation in carrot undergoing different drying processes[J]. Journal of Food Science, 2005, 70(8): s520-s526.
[18]	于海明, 李海源, 张欣悦, 等. 水稻秸秆营养穴盘微波热风耦合干燥动力学模型研究[J]. 农业机械学报, 2020, 51(5):339-348.
	YU H M, LI H Y, ZHANG X Y, et al. Drying kinetic model of microwave coupled with hot air drying of straw-based nutrient seedling-growing bowl tray[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(5):339-348. (in Chinese with English abstract)
[19]	PESEK C A, WARTHESEN J J. Photodegradation of carotenoids in a vegetable juice system[J]. Journal of Food Science, 1987, 52(3): 744-746.
[20]	GOLDMAN M, HOREV B, SAGUY I. Decolorization of β-carotene in model systems simulating dehydrated foods. mechanism and kinetic principles[J]. Journal of Food Science, 1983, 48(3): 751-754.
[21]	DAVEY M W, MELLIDOU I, KEULEMANS W. Considerations to prevent the breakdown and loss of fruit carotenoids during extraction and analysis in Musa[J]. Journal of Chromatography A, 2009, 1216(30): 5759-5762.
[22]	CHANDLER L A, SCHWARTZ S J. Isomerization and losses of trans-.beta.-carotene in sweet potatoes as affected by processing treatments[J]. Journal of Agricultural and Food Chemistry, 1988, 36(1): 129-133.
[23]	ACHIR N, RANDRIANATOANDRO V A, BOHUON P, et al. Kinetic study of β-carotene and lutein degradation in oils during heat treatment[J]. European Journal of Lipid Science and Technology, 2010, 112(3): 349-361.
[24]	HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70(1/2/3): 489-501.
[25]	吴鑫, 宋飞虎, 裴永胜, 等. 基于机器视觉的茶叶微波杀青中品质变化与预测研究[J]. 茶叶科学, 2021, 41(6): 854-864.
	WU X, SONG F H, PEI Y S, et al. Study on the tea quality changes and predictions during the microwave fixation process by machine vision[J]. Journal of Tea Science, 2021, 41(6): 854-864. (in Chinese with English abstract)

南瓜干燥过程中β-胡萝卜素的多源融合预测

Prediction of multi-source fusion of β-carotene during pumpkin drying

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 25

相关文章 15

编辑推荐

Metrics

本文评价

[1]	熊兴伟, 王艺琴, 田怀志, 张素勤, 耿广东. 基于转录组测序解析南瓜子叶黄化的分子机理[J]. 浙江农业学报, 2023, 35(1): 90-102.
[2]	丁燕玲, 王鹏飞, 杨朝云, 周小南, 赵志艳, 张岩峰, 史远刚, 康晓龙. 牛miR-144靶基因预测与组织表达分析[J]. 浙江农业学报, 2022, 34(3): 471-479.
[3]	闫宁, 张晗, 董宏图, 康凯, 罗斌. 基于透射光和反射光图像同位分割的小麦品种识别方法研究[J]. 浙江农业学报, 2022, 34(3): 590-598.
[4]	吴燕君, 洪文英, 章忠梅, 吴耀, 缪强. 设施黄瓜白粉病流行动态与预测模型[J]. 浙江农业学报, 2022, 34(1): 104-111.
[5]	吉训生, 姜晓卫, 夏圣奎. 基于LSTM-Kalman模型的蛋鸡产蛋率预测方法[J]. 浙江农业学报, 2021, 33(9): 1730-1739.
[6]	陈雯, 张伟伟, 邵淑丽, 付学鹏, 黄鑫, 李铁. miR-423-5p在牛肌肉组织中表达及其靶基因预测[J]. 浙江农业学报, 2021, 33(5): 785-793.
[7]	龙鸣, 张棚, 刘元林, 摆小琴, 张福梅, 田晓静, 曹竑, 周雪雁, 马忠仁, 罗丽, 宋礼, Nurul Izza NORDIN. 基于小鼠粪便气味信息的牛乳清蛋白功能效果的鉴别区分[J]. 浙江农业学报, 2021, 33(4): 714-723.
[8]	孟司奇, 丁高蓝, 肖兴宁, 肖英平, 杨华, 汪雯. 巴氏杀菌乳中金黄色葡萄球菌的生长预测模型构建[J]. 浙江农业学报, 2020, 32(7): 1281-1288.
[9]	樊慧丽, 付文阁. 基于局部均衡模型的中国羊肉市场供需变动分析[J]. 浙江农业学报, 2020, 32(6): 1123-1132.
[10]	郭富城, 金丽, 苏强, 粟雨芯, 李方琳, 陈士恩, 马晓霞. PEDV N蛋白的原核表达纯化及B细胞抗原表位预测分析[J]. 浙江农业学报, 2020, 32(5): 762-769.
[11]	包晓敏, 盛家文. 粘虫板害虫自动识别计数研究[J]. 浙江农业学报, 2019, 31(9): 1516-1522.
[12]	马露, 孔维洲, 王丽萍, 张喜康, 王聪, 李佩佩, 刘敦华. 响应面法优化发酵鹿肉干工艺及其货架期预测[J]. 浙江农业学报, 2019, 31(12): 2109-2119.
[13]	田晓静, 龙鸣, 王俊, 马忠仁, 韦真博, 陈士恩, 高丹丹, 丁波. 基于电子鼻气味信息和多元统计分析的枸杞子产地溯源研究[J]. 浙江农业学报, 2018, 30(9): 1604-1611.
[14]	杨煜岑, 杨联安, 任丽, 李聪莉, 朱群娥, 王天泰, 李新尧. 基于随机森林的农耕区土壤有机质空间分布预测[J]. 浙江农业学报, 2018, 30(7): 1211-1217.
[15]	江叶枫, 郭熙. 基于辅助变量和回归径向基函数神经网络(R-RBFNN)的土壤有机质空间分布模拟[J]. 浙江农业学报, 2018, 30(4): 640-648.