基于机器视觉水果分拣系统研究  

Research on fruit sorting system based on machine vision

作　　者：许虎 惠宇龙 万宏强[2] XU Hu;HUI Yulong;WAN Hongqiang(Xi’an Guoxin IoT Technology Co.,Ltd.,Xi’an 710065,China;School of Mechatronic Engineering,Xi’an Technological University,Xi’an 710021,China)

机构地区：[1]西安国信物联技术有限公司,陕西西安710065 [2]西安工业大学机电工程学院,陕西西安710021

出　　处：《现代电子技术》2024年第17期136-142,共7页Modern Electronics Technique

基　　金：西安市2022年科技计划项目(22NYYF063);碑林区2022年科技计划项目(GX2218)。

摘　　要：为解决市面上水果分拣设备体积庞大、效率低等问题,文中给出的水果分拣系统以苹果为测试样本,将相机拍摄的RGB图像转换为HSV图像,并根据H分量分布情况计算苹果表面的色泽度,同时用Canny边缘检测算法提取苹果边缘的轮廓,用最小外接圆法计算苹果果径的大小,结合苹果色泽度和果径大小对苹果进行等级分级。系统试验表明,样本颜色与大小均与苹果的特征相符,分拣设备和人工分拣果径大小误差在±1.35 mm以内。该系统可实现精确分拣、自动化运行等目标,提高了分拣精度及效率。In view of the huge volume and low efficiency of the fruit sorting equipment in the market,a novel fruit sorting system is designed in this paper.In the designed system,the apples are taken as the testing samples.The RGB(red,green,blue)images taken by camera are converted into HSV(hue,saturation,value)images.The grades of the apple colors are calculated according to the H component distribution.The Canny edge detection algorithm is used to extract the apple contours.The minimum circumferential circle method is used to calculate the apple diameters.The apples are graded by the parameters of their color grades and diameters.The system testing shows that the colors and sizes of the samples are consistent with those of the apples,and the error between the fruit diameter obtained by sorting equipment and that by manual sorting is within±1.35 mm.In conclusion,the system can realize the goals of precise sorting and automatic operation,so it can improve the sorting accuracy and efficiency.

关 键 词：机器视觉 水果分拣 HSV颜色模型 CANNY边缘检测算法 轮廓提取 最小外接圆法 

分 类 号：TN911.23-34[电子电信—通信与信息系统] TP23[电子电信—信息与通信工程]