基于机器视觉的白酒包装贴标分拣系统  

Sorting System of Liquor Packaging Labeling Based on Machine Vision

作　　者：李致金[1] 高伟 江凯强 刘忠洋 LI Zhijin;GAO Wei;JIANG Kaiqiang;LIU Zhongyang(School of Artificial Intelligence(School of Future Technology),Nanjing University of Information Science and Technology,Nanjing 210044)

机构地区：[1]南京信息工程大学人工智能学院(未来技术学院),南京210044

出　　处：《计算机与数字工程》2025年第1期45-50,83,共7页Computer & Digital Engineering

基　　金：国家自然科学基金面上项目(编号:62101274)资助。

摘　　要：针对传统白酒包装贴标分拣效率低的问题,文章搭建了基于机器视觉的分拣系统,该系统使用OpenCv开源视觉库设计算法程序,通过工业相机采集包装图像,对图像进行预处理,采用动态阈值分割的方式,提取包装图像边缘特征信息。通过最小外接矩形算法定位图像的中心坐标,实现主商标的ROI区域定位,将商标主要的ROI区域边缘Hu矩特征与主模板的Hu矩特征进行相似性度量,识别包装商标信息。然后通过主商标ROI区域定位相对偏移的辅助ROI区域,根据辅助ROI区域的Hu矩特征与辅助模板匹配结果,结合最小外接矩形的旋转角度和宽高的比值,实现包装360°位姿定位。实验结果表明,该系统响应快速,定位精度达到0.8 mm,角度控制在0.5°以内,能够满足实际的贴标分拣需求。Aiming at the problem of low sorting efficiency of traditional liquor packaging labeling,this paper builds a sorting system based on machine vision.The system uses OpenCv open source visual library to design algorithm programs,collects packag⁃ing images through industrial cameras,preprocesses images,and extracts edge feature information of packaging images by dynamic threshold segmentation.The minimum circumscribed rectangle algorithm is used to locate the center coordinates of the image,and the ROI region of the main trademark is located.The similarity between the Hu moment characteristics of the main ROI edge of the trademark and the Hu moment characteristics of the main template is measured to identify the packaging trademark information.Then,the relatively offset auxiliary ROI region is located through the main trademark ROI region.According to the Hu moment char⁃acteristics of the auxiliary ROI region and the matching results of the auxiliary template,combined with the rotation angle of the min⁃imum circumscribed rectangle and the ratio of width to height,the 360°pose positioning of packaging is realized.The experimental results show that the system has fast response,the positioning accuracy reaches 0.8 mm,and the angle is controlled within 0.5°,which can meet the actual labeling sorting requirements.

关 键 词：机器视觉 定位 HU矩 最小外接矩形 

分 类 号：TP274.3[自动化与计算机技术—检测技术与自动化装置]