基于机器视觉的食品包装检测系统设计  

Design of Food Packaging Detection System Based on Machine Vision

作　　者：李莹[1] 栾秋平[1] LI Ying;LUAN Qiu-ping(Shandong College of Electronic Technology,Jinan 250200,China)

机构地区：[1]山东电子职业技术学院,济南250200

出　　处：《包装工程》2020年第9期210-214,共5页Packaging Engineering

摘　　要：目的为提高食品外包装美观性,确保食品包装质量,基于机器视觉设计一种食品包装检测系统。方法食品包装检测系统主要包括图像获取模块、图像处理和分析模块、输出执行模块等部分。讨论图像处理的关键技术,在传统小波变换的基础上,提出一种改进算法以增强图像特征信息,提高识别率,实现食品包装边缘检测。以污染、飞墨等典型缺陷为例,论述其特征提取方法,包括圆形度、长宽比、灰度标准差等。最后进行实验研究。结果实验结果表明,所述食品包装检测系统的检测精度在99%以上,具有较高的检测准确性。结论基于机器视觉的食品包装检测系统能够满足食品包装需求。The work aims to design a food packaging detection system based on machine vision,in order to improve the appearance and ensure the quality of food packaging.Food packaging detection system mainly included image acquisition module,image processing and analysis module,output execution module and other parts.The key techniques of image processing were discussed.Based on the traditional wavelet transform,an improved algorithm was proposed to enhance the image feature information,improve the recognition rate and realize the edge detection of food packaging.Taking pollution,flying ink and other typical defects as examples,the methods of feature extraction ware discussed,including roundness,aspect ratio,gray standard difference and so on.Finally,an experimental study was carried out.The experimental results showed that,with detection accuracy of more than 99%,the proposed food packaging detection system had relatively high detection accuracy.The food packaging detection system based on the machine vision can meet the requirements of food packaging.

关 键 词：机器视觉 食品包装 小波变换 特征提取 

分 类 号：TB487[一般工业技术—包装工程] TP391[自动化与计算机技术—计算机应用技术]