本文已被：浏览 1020次 下载 0次
投稿时间：2018-11-22
投稿时间：2018-11-22
中文摘要: 利用可见/近红外高光谱（400 nm～1 000 nm）成像技术实现对荷斯坦奶牛、秦川牛、西门塔尔牛、安格斯牛、力木赞牛5 个品种牛肉进行快速无损判别。首先对原始光谱进行预处理，并利用光谱-理化值共生距离法（sample set partitioning based on joint X-Y distance ，SPXY）法划分样本集；结合偏最小二乘判别模型（partial least squaresdiscrimination analysis，PLS-DA），K 最近邻（K-nearest neighbor，KNN）模型和径向基函数-支持向量机（radial basis function-support vector machine，RBF-SVM）模型进行全波段及特征波段判别分析。结果表明，一阶导数（first derivative，FD）法为最优预处理方法；基于RBF-SVM 法所建模型的校正集与预测集准确率分别为100%、99%。可见，基于高光谱成像技术能够获得较好的牛肉品种判别效果。
中文关键词: 可见/近红外 高光谱成像技术 牛肉品种判别 偏最小二乘判别 径向基函数-支持向量机
Abstract:A fast and non-destructive identification for beef varieties of Holstein cows，qinchuan cattle，simmental，Angus，limozan cattle，using visible/near-infrared（400 nm-1 000 nm）hyperspectral technologies was established.Firstly，preprocessing the original spectrum and using joint X-Y distances（SPXY）method to divide the sample.K-nearest neighbor（KNN），partial least squares discrimination analysis（PLS-DA），and radial basis function-support vector machine（RBF-SVM）models discriminant of beef were established，basing on full spectrum and characteristic wavelengths respectively.The results showed that the first derivative（FD）method was the optimal pretreatment method；and the accuracy of the correction set and prediction set of the RBF-SVM models was 100 % and 99 %，respectively.It was confirmed that using hyperspectral imaging technologies could obtain a better recognition effect of beef varieties.
keywords: visible/near infrared hyper spectral imaging technology beef breeds identification partial least squares-discrimination analysis（PLS-DA） radial basis function-support vector machine（RBF-SVM）
文章编号：201920034 中图分类号： 文献标志码：
基金项目:
引用文本：
津ICP备20000603号-1