摘要
为了解决钢厂企业连铸坯信息跟踪的问题,以提高生产智能化水平,搭建了基于机器视觉的连铸坯端面信息码智能识别系统,并重点研究了该系统的识别方法。由于车间环境或连铸坯表面质量等因素对成像问题的影响,连铸坯端面信息码智能识别难度很大,针对现存的字符识别方法难以满足现场实际产线的需求,提出了一种多方向灰度共生矩阵的图像字符识别方法,该方法利用图像像素之间存在的灰度关系进行特征提取,建立了特征模板库,通过特征值比对进行连铸坯端面信息码的实时智能识别,并详细给出了该识别方法的原理及步骤。结果表明,本方法可有效解决连铸坯端面信息码的自动识别问题,提高了钢厂生产效率以及生产自动化和智能化水平。
In order to solve the problem of continous casting billet information tracking in steel mills and improve the level of intelligence production,an intelligent identification system of end face information code of casting billet was established,and the recognition method of the system was studied emphatically.Due to the influence of some factors such as workshop environment or the end face of casting billet itself on the imaging problem,it was very difficult to recognize the information code of the end face of casting billet intelligently.Aiming at the problem that the existing character recognition methods cann′t meet the needs of the actual production line on site,an image recognition method of multi-directional gray level co-occurrence matrix was proposed.This method used the gray level relationship between the image pixels to extract the feature,and the feature template library was established.The real-time intelligent recognition of the information code of the end face of continuous casting billet was carried out through the comparison of the eigenvalue.The principle and steps of the recognition method were given in detail.The experimental results showed that this recognition system can effectively reduce the error rate of information code identification and reduce the recognition time to improve the production efficiency.It can meet the real-time requirements of the production line.
作者
王春梅
黄风山
刘咪
WANG Chun-mei;HUANG Feng-shan;LIU Mi(School of Electrical Engineering, Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China;School of Mechanical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China)
出处
《中国冶金》
CAS
北大核心
2019年第5期33-37,共5页
China Metallurgy
基金
国家自然科学基金资助项目(51075119)
河北省自然科学基金资助项目(E2017208111)
关键词
连铸坯端面
灰度共生矩阵
特征提取
字符识别系统
end face of continuous casting billet
gray level co-occurrence matrix
feature extraction
character identification system
