中国农业科学院农业信息研究所 中国农学会
Agricultural Information Institute of CAAS China Association of Agricultural Science Societies
En
Ch
中国农业科学院农业信息研究所 中国农学会 主办

AI检索

订阅

返回顶部
当前位置: 首页 > 文章 > 基于近红外高光谱成像的冷鲜羊肉表面细菌总数检测 食品工业科技 2014,35 (20) 66-68+81
Position: Home > Articles > Measurement of total viable count on chilled mutton surface based on near-infrared hyperspectral imaging technique Science and Technology of Food Industry 2014,35 (20) 66-68+81

基于近红外高光谱成像的冷鲜羊肉表面细菌总数检测

作  者：
郭中华;郑彩英;金灵
单  位：
宁夏大学物理电气信息学院
关键词：
高光谱成像;冷鲜羊肉;细菌总数;无损检测
摘  要：
细菌总数是反映肉品被污染和腐败状况的重要指标,为寻找快速有效的冷鲜羊肉表面细菌总数无损检测方法,本研究利用近红外高光谱(900~1700nm)成像技术对20d贮藏期内的冷鲜羊肉表面细菌总数进行快速无损检测。由80个样本表面高光谱图像获取目标区域反射光谱,采用多元散射校正和二阶导数相结合(MSC+SD)的方法进行预处理。将用主成分分析法对光谱降维后获得6个特征波长作为输入变量,分别采用偏最小二乘回归(PLS)、误差反向传递人工神经网络(BP-ANN)和径向基函数人工神经网络(RBF-ANN)三种方法建立模型对冷却羊肉表面细菌总数进行预测,均取得较好预测结果,其中,神经网络建模效果优于PLS,预测效果最好的是RBF-ANN模型,相关系数R为0.9988,均方根误差RMSEP为0.2507。结果表明,NIR高光谱图像技术可用于冷鲜羊肉表面细菌总数的快速无损检测。
译  名：
Measurement of total viable count on chilled mutton surface based on near-infrared hyperspectral imaging technique
作  者：
GUO Zhong-hua;ZHENG Cai-ying;JIN Ling;School of Physics and Electronic Information Engineering,Ningxia University;Ningxia Key Laboratory of Intelligent Sensing for Desert Information;
关键词：
hyperspectral imaging;;chilled mutton;;total viable count;;nondestructive detection
摘  要：
Total viable count(TVC) is an important index to reflect the contamination and corruption of meat. In order to look for a quick and efficient detection method of TVC on chilled mutton surface,NIR hyperspectral(900 ~1700nm) imaging technique was applied to the nondestructive detective of TVC on chilled mutton surface,there were 80 samples stored 1 to 20 days in this study. Spectral reflectance curves obtained from the target area of hyperspectral imaging of all samples. Two mixed methods:multiplicative scatter correction adding second derivative(MSC+SD) for pretreatment were used. Then dimensions were reduced by principal component analysis(PCA) to get six characteristic wavelengths as the input variables. Three models were established by using partial least squares(PLS),BP artificial neural network(BP-ANN) and radial basis function artificial neural network(RBF-ANN),all of them had achieved better prediction results,in which the neural network modeling was better than PLS. Overall,the best prediction result was based on the radial basis function artificial neural network(RBF-ANN) model,the correlation coefficient and the root mean square error of prediction were 0.9988 and 0.2507. Therefore,hyperspecctral imaging technique could be used for the nondestructive detection of TVC on chilled mutton surface.

相似文章

导出引用
计量
文章访问数: 4
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊

友情链接Friendly link
农业信息研究所Agricultural information institute 农业智能知识服务平台Agricultural intelligent knowledge service platform 农业学术服务平台Agricultural Knowledge Service System 中国知网CNKI 维普网VIP 万方WANFANG DATA
联系我们Contact us

电话Tell：010-82109652-808

邮箱E-mail：agrijournal@caas.cn

服务Service
关于我们About us 意见建议Opinions and suggest
版权说明Copyright
单位：中国农业科学院农业信息研究所 中国农学会 © 2017中国农业科学院农业信息研究所 ICP备案号：京ICP备09089781号-46