基于机器视觉和近红外光谱的壶瓶枣无损检测  

Nondestructive Detection of Huping Jujube Quality Based on Machine Vision and Near-infrared Diffuse Relectance Spectroscopy Technology

作　　者：薛建新[1] 孙海霞[1] 周靖博[1] 张淑娟[1] 

机构地区：[1]山西农业大学工学院,山西太谷030801

出　　处：《山西农业大学学报（自然科学版）》2012年第6期571-573,共3页Journal of Shanxi Agricultural University(Natural Science Edition)

基　　金：高等学校博士学科点专项科研基金(20101403110003);山西省自然科学基金资助项目(2012011030-3);国家自然科学基金资助项目(31271973)

摘　　要：以壶瓶枣为对象探讨用机器视觉和近红外光谱技术检测壶瓶枣内外品质。通过图像处理技术获取壶瓶枣投影面的边缘提取图像,然后使用最小外接矩形法求得图像的像素点个数,以此求得壶瓶枣投影面的面积。采用MSC对壶瓶枣近红外光谱进行预处理,然后分别采用偏最小二乘法(PLS)、主成分回归(PCR)和偏最小二成支持向量机(LS-SVM)3种建模方式对壶瓶枣可溶性固形物的含量进行预测。结果表明,使用LS-SVM模型获得了最优的预测结果,其预测集的相关系数和均方根误差分别为0.9901和0.328。研究表明,机器视觉结合近红外光谱技术能对壶瓶枣内外品质进行综合检测。The machine vision and near-infrared spectroscopy were used to detect the external and internal quality of Huping Jujube nondestructively. Color images of Huping Jujube samples were captured, the processing of image tech-nology was used for calculating the external rectangular area. MSC method was used to pretreat the near-infrared spec-trum. Then the partial least squares （PLS）, principal component regression （PCR） and lest squares support vector ma-chine （LS-SVM） were used to establish the prediction models of soluble solid content. The results showed that the op-timal LS-SVM model was achieved with correlation coefficient of 0. 9901 and root mean square error of cross validation of 0. 328 for prediction set. Machine vision and near-infrared technology can be a good method to synthetically detect the internal and external quality of Huping Jujube.

关 键 词：机器视觉 近红外光谱 壶瓶枣 检测 

分 类 号：S123[农业科学—农业基础科学] S665