SIA OpenIR  > 数字工厂研究室
玉米籽粒破碎率在线检测装置及方法研究
Alternative TitleResearch on Online Inspection Equipment and Method for Broken Rate of Maize Kernels
杨亮1,2
Department数字工厂研究室
Thesis Advisor王卓
ClassificationS513
Keyword玉米籽粒破碎率在线检测 K-Means均值聚类 玉米籽粒形态特征 智能检测系统 田间实验
Call NumberS513/Y29/2018
Pages90页
Degree Discipline机械工程
Degree Name硕士
2018-05-17
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract本文面向籽粒破碎率在线检测需求,设计研发籽粒破碎率在线检测装置,研究籽粒破碎率在线检测方法,研究内容包括以下几个部分: 第一,针对玉米籽粒破碎率在线检测环境层面的难点,设计研发了一种玉米籽粒破碎率在线检测装置。引入有限元分析方法进行装置应变及应力分析,可实现玉米籽粒在线采集、籽粒单层化,为玉米籽粒在线检测奠定了基础。 第二,针对玉米籽粒破碎率在线检测类别层面的难点,提出了一种基于遗传算法图像增强玉米籽粒胚乳特征提取方法。对采集到的彩色图像从RGB颜色空间,转为Lab颜色空间,并利用K-Means聚类算法,对彩色图像进行分割,可以有效的将彩色图像分割成玉米籽粒主体黄色区域、籽粒尖端和胚乳的白色区域及背景的灰色区域,运用最大类间方差算法(OTSU)优化K-Means聚类算法得到的二值化图像,并运用形态学处理对二值化图像进一步优化,填充孔洞及消除孤立点。 第三,提出了一种基于形态特征的破碎玉米籽粒识别方法。 该方法基于玉米籽粒的面积、面积周长比、短轴长轴比、圆度、矩度等多维形态特征,构建破碎玉米籽粒与完整玉米籽粒的分类器,用于识别破碎玉米籽粒和完整玉米籽粒。 第四,为验证本文所提出方法及装置的可行性和有效性,以玉米籽粒破碎率在线检测装置为基础,搭建了智能检测系统。本文对智能检测系统感知层、分析层及机构执行层的硬件系统搭建进行了介绍,并搭建模拟实验平台,进行田间实验,对玉米籽粒在线检测装置、实验方法和智能检测系统进行实验验证,验证结果表明上述装置、方法及智能检测系统可初步满足田间实验采集数据的需求。
Other AbstractThis paper aims at the on-line detection requirements for the rate of kernel broken, designs and develops an on-line detection device for the rate of kernel broken, and studies the on-line detection method for the rate of kernel broken. The research includes the following sections: Firstly, an on-line detection device for maize kernel broken rate was designed and developed to address the difficulty of on-line detection of maize kernel broken rate at the environmental level. The finite element analysis method was introduced to analyze the strain and stress of the device, which can achieve the on-line collection of maize kernels and the monolayerization of kernels, and laid the foundation for on-line detection of maize kernels. Secondly, based on the difficulties of on-line detection of maize kernel broken at the category level, an image enhancement method based on genetic algorithm for maize kernel endosperm feature extraction was proposed. The collected color image is converted from RGB color space to Lab color space, and the K-Means clustering algorithm is used to segment the color image, and can be effectively divided into the yellow area of the main kernels of maize, the white area of the maize tip and the endosperm, and the gray area of the background. The binary image obtained by the K-Means clustering algorithm is optimized by using the OTSU algorithm, and the morphological processing is used to further optimize the binary image, fill the hole, and eliminate the isolated point. Thirdly, a method for identifying the kernels of broken maize based on morphological characteristics was proposed. The method is based on multi-dimensional morphological characteristics such as area, area perimeter ratio, ratio of short axis to major axis, roundness. And a classifier of broken maize kernels and intact maize kernels is constructed for identifying broken maize kernels and intact maize kernels. Fourthly, in order to verify the feasibility and effectiveness of the proposed method and device, an intelligent detection system was built based on the on-line detection device of maize kernel broken. This paper introduced the hardware system construction of the sensing layer, analysis layer and organization execution layer of the intelligent detection system, and set up a simulation experiment platform and conducting field experiments to use the online inspection equipment, experimental method and intelligent detection system of maize kernels were validated experimentally. The verification results showed that the above equipment, method and the intelligent detection system could initially meet the needs of the field experiment data collection.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/21801
Collection数字工厂研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院大学
Recommended Citation
GB/T 7714
杨亮. 玉米籽粒破碎率在线检测装置及方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2018.
Files in This Item:
File Name/Size DocType Version Access License
玉米籽粒破碎率在线检测及方法研究.pdf(3833KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杨亮]'s Articles
Baidu academic
Similar articles in Baidu academic
[杨亮]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杨亮]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.