摘要
以宁夏滩羊肉为研究对象,利用400~1000 nm可见近红外高光谱对冷鲜羊肉的菌落总数和挥发性盐基氮含量进行新鲜度的检测研究。采集冷鲜滩羊肉表面光谱图像,提取感兴趣区域获取原始光谱数据。剔除由蒙特卡洛检测法所检测出的异常样本,采用理化值共生距离法(SPXY)划分样本的校正集和预测集。先对原始光谱预处理并建立偏最小二乘回归(PLSR)模型,优选最佳预处理方法;后采用主成分回归法(PCR)和支持向量机回归法(SVR)建立模型,优选最佳建模方法。结果表明:光谱数据经过正交信号校正(OSC)预处理后,建立的菌落总数和TVB-N含量预测模型效果较好,其RC分别为0.9067和0.9147,Rp分别为0.8743和0.8802,均高于其他光谱预处理模型。通过不同建模方法的比较,建模效果较好的是PLSR方法。研究表明:利用可见近红外高光谱技术可以实现对滩羊肉新鲜度的无损检测。
The freshness of colony and the content of volatile base nitrogen in cold fresh mutton were studied by using near- infrared hyperspectral spectrum of 400-1000 nm in Ningxia Tan mutton as the research object.The surface spectral images of cold and fresh muttons were collected and the original spectral data were extracted from the region of interest.The anomalous samples detected by the Monte Carlo method were excluded and the calibration set and the prediction set of the samples were divided by the physical and chemical value symbiotic distance method （SPXY） , （PLSR） model and the best pretreatment method was established.Then, the principal component regression （PCR）and support vector machine regression （SVR）were used to establish the model.Modular method.The results showed that the predicted number of colonies and TVB-N were better when the spectral data were pretreated by orthogonal signal correction （ OSC ） , the RCs were 0.9067 and 0.9147, and the Rp were 0.8743 and 0.8802, which were higher than other spectral pretreatment models. Through the comparison of different modeling methods, better PLSR method was the best method. The results showed that Nondestructive testing of freshness of beach mutton could be achieved by using near-infrared hyperspectral spectroscopy.
出处
《食品工业科技》
CAS
CSCD
北大核心
2017年第22期235-241,共7页
Science and Technology of Food Industry
基金
中央财政支持地方高校改革发展资金--食品学科建设项目(2017)
国家自然科学基金资助项目(31660484)
关键词
高光谱成像技术
滩羊肉
菌落总数
挥发性盐基氮
无损检测
偏最小二乘回归
hyperspeetral imaging
Tan- mutton
total number of colonies
volatile base nitrogen
nondestructive testing
partial least squares regression
