当前位置: 首页 > 文章 > 花生籽仁感官品质鉴定中的计算机色选机制研究 花生学报 2009,38 (2) 15-19
Position: Home > Articles > A Study on Color Based Quality Identification Method of Peanut Kernel with Computer Journal of Peanut Science 2009,38 (2) 15-19

花生籽仁感官品质鉴定中的计算机色选机制研究

作  者:
韩仲志;刘竟;杨锦忠
单  位:
青岛农业大学信息科学与工程学院;青岛农业大学植物科技学院;青岛农业大学数字农业研究中心
关键词:
花生仁;图像处理;品质检测;颜色特征;统计检验
摘  要:
花生籽仁的品质鉴定是花生外贸加工和质量检验的重要内容。为了验证基于颜色特征的品质分级的可行性,数码相机微距获取了6大类品质(标准、揉伤、切伤、霉变、发芽、杂质)花生籽仁图像192幅,建立和检验了12个颜色特征及其组合组成的15个特征的识别模型。发现两个及两个以上颜色分量特征的识别率已达到90%以上,采用三个颜色分量基本达到了实用的要求。机器视觉检测具有成本和速度上的优势,能够用于花生籽仁的品质鉴定,G1、R1、R2组合模型最佳,分辨率达到了实用的要求。经改进技术识别率可进一步提高。
译  名:
A Study on Color Based Quality Identification Method of Peanut Kernel with Computer
作  者:
HAN Zhong-zhi1,2,LIU Jing2,YANG Jin-Zhong1,3(1.Digital Agriculture Research Center,Qingdao Agri.Uni.,Qingdao 266109,China;2.College of Information Sci.& Engineering,Qingdao Agri.Uni.,Qingdao 266109,China;3.Plant Science & Technology College,Qingdao Agri.Uni.,Qingdao 266109,China)
关键词:
peanuts;image process;quality identification;seed color;stat.analysis
摘  要:
Seed quality plays a crucial role in foreign trade and seed quality testing in peanut.To examine the feasibility of quality classification by color characteristics,we obtained 192 pictures of 6 kinds of different quality by macroshot of digital camera.They are normal,knead damaged,knives damaged,moldy,germinated peanuts and some impurities.15 recognition models were created by 12 colors characteristics,and which has been tested.We found that two and plural color characteristic components can obtain recognize rate of higher than 90%,three colors characteristics already can be used practicality.Machine vision wins the advantages of low cost and high speed over manual or biochemical detecting methods,and is feasible to be applied to identification of peanuts quality.The combination of R1,R2 and G1 is the best model.Model performance may be promoted further with optimizing samples and structure.

相似文章

计量
文章访问数: 10
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊