基于形态特征的破碎玉米籽粒识别及检测装置设计  

Recognition of the Broken Corn Kernels Based on Morphology Feature And Design of Detection Device

作　　者：杨亮[1,2,3] 王卓[1,2,3] 白晓平 Yang Liang;Wang Zhuo;Bai Xiaoping(Shenyang Institute of Automation Chinese Academy of Sciences,Shenyang 110016,China;Institutes of Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang 100169,China;Key Laboratory of Agricultural Equipment Intelligent Technology in Liaoning Province,Shenyang 110016,China)

机构地区：[1]中国科学院沈阳自动化研究所,沈阳110016 [2]中国科学院机器人与智能制造创新研究院,沈阳110169 [3]辽宁省农业装备智能化技术重点实验室,沈阳110016

出　　处：《农机化研究》2024年第11期21-28,共8页Journal of Agricultural Mechanization Research

基　　金：辽宁省重点研发计划项目(2020JH2/10200026);国家重点研发计划项目(2020YFB1709600)。

摘　　要：基于K-Means聚类分割算法、颜色空间转换和形态学特征,通过对辽沈地区广泛种植的玉米的面积、面积与周长比、短轴与长轴比、圆形与矩形及五种形态特征的分析,识别出玉米碎粒。同时,设计开发了一种基于同步带轮的玉米籽粒破碎率在线检测装置,用于玉米籽粒破碎率的在线检测与识别。研究结果:①利用K平均算法分割算法将图像分割为目标区域和背景区域;②利用K平均算法分割算法将图像分割为二值化图像,进一步对二值化图像采用形态闭运算填充目标区域的非连通区域,对目标区域进行统计分析,计算出形态特征;③通过多组试验,玉米籽粒破碎的识别率可达94%。Study on the K-Means clustering segmentation algorithm,color space conversion and morphology feature.Analyzing the corns’area,the ratio of area and perimeter,the ratio of short axis and long axis,roundness and rectangularity,five morphology features to recognize the broken corn kernels that they are widely planted in Liao-Shen Area.At the same time,an online detection device for corn kernels broken rate based on synchronous pulley is designed and developed for online detection and recognition of corn kernels broken rate.Results:①using the K-Means clustering segmentation algorithm to divide the images into the target region and the background region;②The images have divided by K-Means clustering segmentation algorithm convert to the binarized images,further the binarized images use the morphological close operation to fill these are not connected areas of the target region,statistical analysis of the target region and we calculate the morphology features;③Through many groups of experiments,the recognition rate of corn kernels breakage can reach up to 94%.

关 键 词：玉米籽粒 K-Means聚类分割 形态特征 颜色空间 在线检测装置 

分 类 号：S225.51[农业科学—农业机械化工程] S220.3[农业科学—农业工程]