首页 | 本学科首页   官方微博 | 高级检索  
     检索      

猕猴桃自动分级设备设计与试验
引用本文:左兴健,武广伟.猕猴桃自动分级设备设计与试验[J].农业机械学报,2014,45(S1):287-295.
作者姓名:左兴健  武广伟
作者单位:北京农业智能装备技术研究中心;北京农业智能装备技术研究中心
基金项目:新疆生产建设兵团重大科技资助项目(2013AA004-2)
摘    要:针对目前猕猴桃采后人工分级费时费力、自动分级成本高等问题,通过机械分离输送、图像采集与处理、智能控制等技术,研制了一套实用的猕猴桃自动分级设备。该设备包括单行定位输送系统、图像采集系统、分级执行系统和控制系统,通过多特征提取和融合分级的方法对猕猴桃实现了自动分级,同时在线监控参数。经试验验证,该设备按体积、形状、表面缺陷特征分级的准确率分别可达88.9%、91%、94%,融合分级的准确率可达86%。

关 键 词:猕猴桃  分级  图像处理  BP神经网络
收稿时间:7/7/2014 12:00:00 AM

Design and Experiment on Automatic Grading Machine for Kiwi
Zuo Xingjian and Wu Guangwei.Design and Experiment on Automatic Grading Machine for Kiwi[J].Transactions of the Chinese Society of Agricultural Machinery,2014,45(S1):287-295.
Authors:Zuo Xingjian and Wu Guangwei
Institution:Beijing Research Center for Intelligent Agriculture Equipments;Beijing Research Center for Intelligent Agriculture Equipments
Abstract:The kiwifruit automatic grading machine is an important machine for postharvest processing of kiwifruit. A set of practical automatic kiwi fruit grading machine was designed based on technologies of mechanical separation and conveyance, image capture and processing as well as intelligent control. This set of equipment mainly consisted of material storage bin, single line positioning system, image capture system and grading executive system. And it also has programs of image capture, image processing, grade determination and human-machine interaction for the host computer as well as control programs for the slave computer. By testing, this machine was capable of grading by three features of size, shape and surface defect and their grading accuracy could reach to 88.9%, 91% and 94%, respectively. And the accuracy of integrated grading could reach to 86%.
Keywords:Kiwi  Grading  Image processing  BP neural network
点击此处可从《农业机械学报》浏览原始摘要信息
点击此处可从《农业机械学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号