基于随机蛙跳和支持向量机的牛乳收购分级模型构建

doi:10.12133/j.smartag.2021.3.4.202107-SA003

摘要/Abstract

摘要：

蛋白质、脂肪含量和体细胞数量作为牛乳收购中的重要参考指标，决定了牛乳的品质和价格。为批量准确地对牛乳品质进行分级，提高乳企的生产效率，本研究以3216份荷斯坦牛牛乳样本为研究对象，应用中红外光谱技术实现对收购过程中4种不同品质牛乳的检测分级。利用一阶导数和一阶差分对光谱进行预处理，并结合竞争性自适应重加权算法（Competitive Adaptive Reweighted Sampling，CARS）和随机蛙跳算法（Shuffled Frog Leaping Algorithm，SFLA）筛选出能代表不同牛乳的有效特征变量，建立支持向量机（Support Vector Machine，SVM）模型。其中，利用网格搜索法（Grid Search，GS）、遗传算法（Genetic Algorithm，GA）和粒子群算法（Particle Swarm Optimization，PSO）对SVM模型的关键参数——惩罚参数c和核函数参数g进行优化。结果表明，SFLA算法总体上优于CARS算法，PSO优化SVM模型的效果最佳。一阶差分预处理后，利用SFLA算法筛选特征变量建立的PSO-SVM模型，训练集准确率、测试集准确率和曲线下面积（Area Under Curve，AUC）分别为97.8%、95.6%和0.96489。该模型具有较高的准确率，在牛乳产业中具有实际应用价值。

关键词: 中红外光谱, 牛乳, 收购分级, 随机蛙跳, 支持向量机

Abstract:

Protein, fat and somatic cells are three important reference indicators in milk purchase, which determine the quality and price of milk. The traditional chemical analysis methods of these indexes are time-consuming and pollute the environment, while the mid-infrared spectrum has the advantages of fast, non-destructive and simple operation. In order to realize the rapid classification of milk quality and improve the production efficiency of dairy enterprises, 3216 Holstein milk samples were chosen as the research objects and mid-infrared spectroscopy technology was applied to realize the detection and classification of 4 different quality milks during the purchase process. The spectrum was preprocessed by using the first derivative and the first difference, and combined with the algorithm competitive adaptive reweighted sampling (CARS) and the shuffled frog leaping algorithm (SFLA), the effective characteristic variables that could represent different milks were selected, and the SVM model was established. Among them, the penalty parameter c and the kernel function parameter g which were the key parameters of the SVM model were optimized by using the grid search method (GS), genetic algorithm (GA) and particle swarm algorithm (PSO). The training time of GS, GA and PSO algorithms were compared, the results showed that the training time of GS was much longer than that of GA and PSO algorithms.The SFLA algorithm was generally better than the CARS algorithm, and the PSO optimized the SVM model the best. After the first-order difference preprocessing, the PSO-SVM established by using the SFLA algorithm to filter the characteristic variables, the accuracy of the training set, the accuracy of the test set and the AUC were 97.8%, 95.6% and 0.96489, respectively. This model has a high accuracy rate and has practical application value in the milk industry.

Key words: mid-infrared spectrum, milk, purchase classification, shuffled frog leaping algorithm, support vector machine

中图分类号:

肖仕杰, 王巧华, 李春芳, 赵利梅, 刘鑫雅, 卢士宇, 张淑君. 基于随机蛙跳和支持向量机的牛乳收购分级模型构建[J]. 智慧农业(中英文), 2021, 3(4): 77-85.

XIAO Shijie, WANG Qiaohua, LI Chunfang, ZHAO Limei, LIU Xinya, LU Shiyu, ZHANG Shujun. Construction of Milk Purchase Classification Model Based on Shuffled Frog Leaping Algorithm and Support Vector Machine[J]. Smart Agriculture, 2021, 3(4): 77-85.

参考文献

1	朱海明, 程启方. 瑞典牛奶检测分级付款系统简介[J]. 中国奶牛, 1997(4): 52-54.
1	ZHU H, CHENG Q. Brief introduction of Swedish milk testing grading payment system[J]. China Dairy Cattle, 1997(4): 52-54.
2	史慧茹, 姜瞻梅, 田波. 牛乳体细胞数的检测方法[J]. 畜牧与饲料科学, 2008(2): 86-88.
2	SHI H, JIANG Z, TIAN B. Method for detecting somatic cell count in bovine milk[J]. Animal Husbandry and Feed Science, 2008(2): 86-88.
3	陈贺, 王帅, 陈红玲. 乌鲁木齐地区生鲜牛乳质量分级研究[J]. 农村科技, 2017(8): 60-62.
3	CHEN H, WANG S, CHEN H. Study on the quality classification of fresh milk in Urumqi area[J]. Rural Science & Technology, 2017(8): 60-62.
4	SMITH K L. Standards for somatic cells in milk: Physiologicaland regulatory[J]. IDF Mastitis Newslett, 1995, 144 (21): 7-9.
5	KOLDWIJ E, EMANWLSON U. Relation of milk production lossto milk somatic cell count[J]. ACTA Vet Scand, 1999, 40: 47-56.
6	GONDIM C, JUNQUEIRA R G, VITORINO C D S S, et al. Detection of several common adulterants in raw milk by MID-infrared spectroscopy and one-class and multi-class multivariate strategies[J]. Food Chemistry, 2017, 230: 68-75.
7	TOFFANIN, V, PENASA, M, MCPARLAND, S, et al. Genetic parameters for milk mineral content and acidity predicted by mid-infrared spectroscopy in Holstein-Friesian cows[J]. Animal, 2015, 9(5): 775-780.
8	SOYEURT H, DEHARENG F, GENGLER N, et al. Mid-infrared prediction of bovine milk fatty acids across multiple breeds, production systems, and countries[J]. Journal of Dairy Science, 2011, 94(4): 1657-1667.
9	李巧玲, 刘峰, 宋思远, 等. 中红外光谱法快速测定牛奶中非蛋白氮类物质[J]. 食品工业科技, 2014, 35(22): 73-75, 80.
9	LI Q, LIU F, SONG S, et al. Fast determination of nonprotein nitrogen content in milk based on mid-infrared spectroscopy method[J]. Science and Technology of Food Industry, 2014, 35(22): 73-75, 80.
10	吴珽, 梁龙, 朱华, 等. 海南制浆树种中主要成分的近红外分析与模型优化[J]. 光谱学与光谱分析, 2021, 41(5): 1404-1409.
10	WU T, LIANG L, ZHU H, et al. Near-infrared analysis and models optimization of main components in Pulpwood of Hainan province[J]. Spectroscopy and Spectral Analysis, 2021, 41(5): 1404-1409.
11	花晨芝, 赵凌, 宋建军, 等. 粒子群算法选择特征波长在紫外光谱检测COD中的研究[J]. 西华师范大学学报(自然科学版), 2019, 40(1): 81-85.
11	HUA C, ZHAO L, SONG J, et al. Selection of wavelength for UV-visible spectroscopy based on BLS combined with PSO[J]. Journal of China West Normal University (Natural Sciences), 2019, 40(1): 81-85.
12	石吉勇, 邹小波, 王开亮, 等. 模拟退火算法用于食醋总酸含量近红外光谱模型的波数点优选[J]. 食品科学, 2011, 32(10): 120-123.
12	SHI J, ZOU X, WANG K, et al. Simulated annealing algorithm based wavenumber selection for total acid content analysis in vinegar by near infrared spectroscopy[J]. Food Science, 2011, 32(10): 120-123.
13	刘冬阳, 孙晓荣, 刘翠玲, 等. 拉曼光谱结合模拟退火的小麦粉灰分含量检测[J]. 中国粮油学报, 2019, 34(5): 128-133.
13	LIU D, SUN X, LIU C, et al. Detection of ash control of wheat flour based on Raman spectroscopy combined with simulated annealing[J]. Journal of the Chinese Cereals and Oils Association, 2019, 34(5): 128-133.
14	周孟然, 孙磊, 卞凯, 等. iPLS波段筛选方法在食用油品上快速检测研究[J]. 激光杂志, 2020, 41(7): 13-17.
14	ZHOU M, SUN L, BIAN K, et al. Band screening of iPLS for laser-induced fluorescence spectrum of edible oil[J]. Laser Journal, 2020, 41(7): 13-17.
15	张烝彦, 叶沁, 刘晓颖, 等. 傅里叶变换衰减全反射红外光谱结合向前区间偏最小二乘法快速测定食用油中总极性化合物[J]. 浙江农业科学, 2019, 60(6): 1003-1007.
15	ZHANG Z, YE Q, LIU X, et al. Fourier transform attenuated total reflection infrared spectroscopy combined with forward interval partial least squares method for rapid determination of total polar compounds in edible oil[J]. Journal of Zhejiang Agricultural Sciences, 2019, 60(6): 1003-1007.
16	王拓, 戴连奎, 马万武. 拉曼光谱结合后向间隔偏最小二乘法用于调和汽油辛烷值定量分析[J]. 分析化学, 2018, 46(4): 623-629.
16	WANG T, DAI L, MA W. Quantitative analysis of blended gasoline octane number using Raman spectroscopy with backward interval partial least squares method[J]. Chinese Journal of Analytical Chemistry, 2018, 46(4): 623-629.
17	史智佳, 李鹏飞, 吕玉, 等. 移动窗口偏最小二乘法优选猪油丙二醛近红外光谱波段[J]. 中国食品学报, 2014, 14(11): 207-213.
17	SHI Z, LI P, LYU Y, et al. Region optimization in FT-NIR spectroscopy for determination of MDA in lard with moving window partial least squares[J]. Journal of Chinese Institute of Food Science and Technology, 2014, 14(11): 207-213.
18	许良, 闫亮亮, 塞击拉呼, 等. 近红外光谱结合可移动窗口偏最小二乘法对克霉唑粉末药品的定量分析[J]. 计算机与应用化学, 2016, 33(4): 415-418.
18	XU L, YAN L, SAIJLAHU, et al. Quantitative analysis of Clotrimazole powder drugs by using moving window partial least square method combined with near-infrared spectroscopy[J]. Computers and Applied Chemistry, 2016, 33(4): 415-418.
19	李庆旭, 王巧华, 马美湖, 等. 基于可见/近红外光谱和深度学习的早期鸭胚雌雄信息无损检测[J]. 光谱学与光谱分析, 2021, 41(6): 1800-1805.
19	LI Q, WANG Q, MA M, et al. Non-destructive detection of male and female information of early duck embryos based on visible/near infrared spectroscopy and deep learning[J]. Spectroscopy and Spectral Analysis, 2021, 41(6): 1800-1805.
20	付丹丹, 王巧华, 高升, 等. 不同品种鸡蛋贮期S-卵白蛋白含量分析及其可见/近红外光谱无损检测模型研究[J]. 分析化学, 2020, 48(2): 289-297.
20	FU D, WANG Q, GAO S, et al. Analysis of S-Ovalbumin content of different varieties of eggs during storage and its nondestructive testing model by visible-near infrared spectroscopy[J]. Chinese Journal of Analytical Chemistry, 2020, 48(2): 289-297.
21	韩毅, 蔡建湖, 周根贵, 等. 随机蛙跳算法的研究进展[J]. 计算机科学, 2010, 37(7): 16-19.
21	HAN Y, CAI J, ZHOU G, et al. Advances in shuffled frog leaping algorithm[J]. Computer Science, 2010, 37(7): 16-19.
22	孙晶京, 杨武德, 冯美臣, 等. 基于随机蛙跳和支持向量机的冬小麦叶面积指数估算[J]. 山西农业大学学报(自然科学版), 2020, 40(5): 120-128.
22	SUN J, YANG W, FENG M, et al. Estimation of winter wheat leaf area index based on random leapfrog and support vector regression approach[J]. Journal of Shanxi Agricultural University (Natural Science Edition), 2020, 40(5): 120-128.
23	王巧华, 梅璐, 马美湖, 等.利用机器视觉与近红外光谱技术的皮蛋无损检测与分级[J]. 农业工程学报, 2019, 35(24): 314-321..
23	WANG Q, MEI L, MA M, et al.Nondestructive testing and grading of preserved duck eggs based on machine vision and near-infrared spectroscopy[J]. Transactions of the CSAE, 2020, 40(5): 120-128.
24	黄平捷, 李宇涵, 俞巧君, 等. 基于SPA和多分类SVM的紫外-可见光光谱饮用水有机污染物判别方法研究[J]. 光谱学与光谱分析, 2020, 40(7): 2267-2272.
24	HUANG P, LI Y, YU Q, et al. Classify of organic contaminants in water distribution systems developed by SPA and multi-classification SVM using UV-VIS spectroscopy[J]. Spectroscopy and Spectral Analysis, 2020, 40(7): 2267-2272.
25	Vapnik V N. An overview of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1999, 10(10): 988-999.
26	Burges C J C. A Tutorial on support vector machines for pattern recognition[J]. Data Mining and Knowledge Discovery. 1998, 2(2): 121-167.
27	BONFATTI V, MARTINO G D, CARNIER P. Effectiveness of mid-infrared spectroscopy for the prediction of detailed protein composition and contents of protein genetic variants of individual milk of Simmental cows[J]. Journal of Dairy Science, 2010, 94(12): 5776-5785.
28	代芬, 邱泽源, 邱倩, 等. 基于拉曼光谱和自荧光光谱的柑橘黄龙病快速检测方法[J]. 智慧农业, 2019, 1(3): 77-86.
28	DAI F, QIU Z, QIU Q, et al. Rapid detection of citrus Huanglongbing using Raman spectroscopy and auto-fluorescence spectroscopy[J]. Smart Agriculture, 2019, 1(3): 77-86.
29	胡翼然, 李杰庆, 刘鸿高, 等. 基于支持向量机对云南常见野生食用牛肝菌中红外光谱的种类鉴别[J]. 食品科学, 2021, 42(8): 248-256.
29	HU Y, LI J, LIU H, et al. Species identification of common wild edible bolete in Yunnan by Fourier transform mid-infrared spectroscopy coupled with support vector machine[J]. Food Science, 2021, 42(8): 248-256.

[1]	魏永康, 杨天聪, 丁信尧, 高越之, 袁鑫茹, 贺利, 王永华, 段剑钊, 冯伟. 基于不同空间分辨率无人机多光谱遥感影像的小麦倒伏区域识别方法[J]. 智慧农业(中英文), 2023, 5(2): 56-67.
[2]	赖佳政, 李贝贝, 程翔, 孙丰, 陈炬廷, 王晶, 张芊, 叶协锋. 基于无人机高光谱遥感的烤烟叶片叶绿素含量估测[J]. 智慧农业(中英文), 2023, 5(2): 68-81.
[3]	赵晋陵, 杜世州, 黄林生. 联合多源多时相卫星影像和支持向量机的小麦白粉病监测方法[J]. 智慧农业(中英文), 2022, 4(1): 17-28.
[4]	李东博, 黄铝文, 赵旭博. 基于介电特征的苹果霉心病检测方法[J]. 智慧农业(中英文), 2021, 3(4): 66-76.
[5]	刘恬琳, 朱西存, 白雪源, 彭玉凤, 李美炫, 田中宇, 姜远茂, 杨贵军. 土壤有机质含量高光谱估测模型构建及精度对比[J]. 智慧农业（中英文）, 2020, 2(3): 129-138.
[6]	FLORESPaulo, MATHEWJithin, JAHANNusrat, STENGERJohn. 利用图像和机器学习检测大豆作物幼苗期玉米杂苗[J]. 智慧农业（中英文）, 2020, 2(3): 61-74.
[7]	周超, 徐大明, 吝凯, 陈澜, 张松, 孙传恒, 杨信廷. 基于近红外机器视觉的鱼类摄食强度评估方法研究[J]. 智慧农业, 2019, 1(1): 76-84.