本文主要研究内容
作者魏文松,彭彦昆,郑晓春,王文秀(2019)在《基于多光谱漫反射的牛肉品质参数检测方法研究》一文中研究指出:为了满足生鲜肉品质参数无损检测领域,对轻便式、低成本设备的开发需求,提出一种基于多光谱漫反射技术的生鲜肉品质检测方法。首先根据漫反射近似理论,结合牛肉样品散射系数、吸收系数及折射率等参数,在无线细垂直光束的蒙特卡洛仿真的基础上,对具有一定发散角度LED光源进行了初始化的校正,分别从光源照射位置概率分布、不同角度的照射概率分布、仰角、方向角的概率分布、不同角度光线入射样品时反射引起能量损失及对光子权重的影响,得到在LED光源发散角情况下,不同源探距下的漫反射率与检测深度,确定了光源与检测器之间的最佳距离为15 mm,然后根据此距离,搭建了多光谱漫反射检测平台,检测平台由8组中心波长为470, 535, 575, 610, 650, 720, 780和960 nm的LED光源组成,与所要检测的生鲜牛肉品质参数相对应。同时利用LED光源的发散角,确定了光源到样品表面的垂直距离与每个光源的安装位置,保证光源照射到样品的区域是均匀的。样品的漫射光强经由信号采集与放大电路的处理后传至上位机,并在上位机完成建模与分析。最后为验证该检测系统的性能,以生鲜牛肉新鲜度参数中的颜色(L~*, a~*, b~*)与pH值为指标,利用60个样品进行了试验,分别得到8个光源下的原始光强值与校正后的反射率值,然后将牛肉样品按照3∶1比例分为校正集与预测集,针对原始光强值与反射率值,分别利用多元线性回归(multiple linear regression, MLR),偏最小二乘回归(partial least squares regression, PLSR)与偏最小二乘支持向量机回归(partial least-squares support vector machine, LS-SVM)三种方法,建立各个参数在原始光强与反射率数据两种情况下的预测模型,并得到最佳模型结果。结果表明,利用反射率数据建模结果均好于光强数据结果,其中参数L~*, a~*, b~*的MLR建模结果优于PLSR与LS-SVR,其预测集相关系数分别为0.983 2, 0.907 2及0.935 9,预测集误差分别为1.00, 2.14及0.67。参数pH值的LS-SVR建模结果优于PLSR与MLR,其预测集相关系数为0.942 0,误差为0.19。最后利用未参与试验的20块牛肉样品对模型进行了验证,颜色L~*, a~*, b~*及pH参数的预测值与实测值的相关系数均大于0.85,结果证明,利用多光谱漫反射技术以及所搭建的多光谱漫反射检测系统对生鲜牛肉品质参数检测是可行的,该方法能够为设计便携式或微型化生鲜牛肉品质的无损检测仪器提供参考与依据。
Abstract
wei le man zu sheng xian rou pin zhi can shu mo sun jian ce ling yu ,dui qing bian shi 、di cheng ben she bei de kai fa xu qiu ,di chu yi chong ji yu duo guang pu man fan she ji shu de sheng xian rou pin zhi jian ce fang fa 。shou xian gen ju man fan she jin shi li lun ,jie ge niu rou yang pin san she ji shu 、xi shou ji shu ji she she lv deng can shu ,zai mo xian xi chui zhi guang shu de meng te ka luo fang zhen de ji chu shang ,dui ju you yi ding fa san jiao du LEDguang yuan jin hang le chu shi hua de jiao zheng ,fen bie cong guang yuan zhao she wei zhi gai lv fen bu 、bu tong jiao du de zhao she gai lv fen bu 、yang jiao 、fang xiang jiao de gai lv fen bu 、bu tong jiao du guang xian ru she yang pin shi fan she yin qi neng liang sun shi ji dui guang zi quan chong de ying xiang ,de dao zai LEDguang yuan fa san jiao qing kuang xia ,bu tong yuan tan ju xia de man fan she lv yu jian ce shen du ,que ding le guang yuan yu jian ce qi zhi jian de zui jia ju li wei 15 mm,ran hou gen ju ci ju li ,da jian le duo guang pu man fan she jian ce ping tai ,jian ce ping tai you 8zu zhong xin bo chang wei 470, 535, 575, 610, 650, 720, 780he 960 nmde LEDguang yuan zu cheng ,yu suo yao jian ce de sheng xian niu rou pin zhi can shu xiang dui ying 。tong shi li yong LEDguang yuan de fa san jiao ,que ding le guang yuan dao yang pin biao mian de chui zhi ju li yu mei ge guang yuan de an zhuang wei zhi ,bao zheng guang yuan zhao she dao yang pin de ou yu shi jun yun de 。yang pin de man she guang jiang jing you xin hao cai ji yu fang da dian lu de chu li hou chuan zhi shang wei ji ,bing zai shang wei ji wan cheng jian mo yu fen xi 。zui hou wei yan zheng gai jian ce ji tong de xing neng ,yi sheng xian niu rou xin xian du can shu zhong de yan se (L~*, a~*, b~*)yu pHzhi wei zhi biao ,li yong 60ge yang pin jin hang le shi yan ,fen bie de dao 8ge guang yuan xia de yuan shi guang jiang zhi yu jiao zheng hou de fan she lv zhi ,ran hou jiang niu rou yang pin an zhao 3∶1bi li fen wei jiao zheng ji yu yu ce ji ,zhen dui yuan shi guang jiang zhi yu fan she lv zhi ,fen bie li yong duo yuan xian xing hui gui (multiple linear regression, MLR),pian zui xiao er cheng hui gui (partial least squares regression, PLSR)yu pian zui xiao er cheng zhi chi xiang liang ji hui gui (partial least-squares support vector machine, LS-SVM)san chong fang fa ,jian li ge ge can shu zai yuan shi guang jiang yu fan she lv shu ju liang chong qing kuang xia de yu ce mo xing ,bing de dao zui jia mo xing jie guo 。jie guo biao ming ,li yong fan she lv shu ju jian mo jie guo jun hao yu guang jiang shu ju jie guo ,ji zhong can shu L~*, a~*, b~*de MLRjian mo jie guo you yu PLSRyu LS-SVR,ji yu ce ji xiang guan ji shu fen bie wei 0.983 2, 0.907 2ji 0.935 9,yu ce ji wu cha fen bie wei 1.00, 2.14ji 0.67。can shu pHzhi de LS-SVRjian mo jie guo you yu PLSRyu MLR,ji yu ce ji xiang guan ji shu wei 0.942 0,wu cha wei 0.19。zui hou li yong wei can yu shi yan de 20kuai niu rou yang pin dui mo xing jin hang le yan zheng ,yan se L~*, a~*, b~*ji pHcan shu de yu ce zhi yu shi ce zhi de xiang guan ji shu jun da yu 0.85,jie guo zheng ming ,li yong duo guang pu man fan she ji shu yi ji suo da jian de duo guang pu man fan she jian ce ji tong dui sheng xian niu rou pin zhi can shu jian ce shi ke hang de ,gai fang fa neng gou wei she ji bian xie shi huo wei xing hua sheng xian niu rou pin zhi de mo sun jian ce yi qi di gong can kao yu yi ju 。
论文参考文献
论文详细介绍
论文作者分别是来自光谱学与光谱分析的魏文松,彭彦昆,郑晓春,王文秀,发表于刊物光谱学与光谱分析2019年04期论文,是一篇关于多光谱漫反射检测论文,蒙特卡洛仿真论文,牛肉品质参数论文,光源系统论文,预测模型论文,光谱学与光谱分析2019年04期论文的文章。本文可供学术参考使用,各位学者可以免费参考阅读下载,文章观点不代表本站观点,资料来自光谱学与光谱分析2019年04期论文网站,若本站收录的文献无意侵犯了您的著作版权,请联系我们删除。
标签:多光谱漫反射检测论文; 蒙特卡洛仿真论文; 牛肉品质参数论文; 光源系统论文; 预测模型论文; 光谱学与光谱分析2019年04期论文;