摘要
为了解决特钢棒材信息跟踪的问题,提高生产智能化水平,搭建了基于机器视觉的特钢棒材端面信息码识别系统。根据车间环境和特钢棒材端面特征,提出了水平双标志点的标记方案,并针对该标记方案提出了一种信息码图像识别方法。该识别方法提取了字符二值图像灰度共生矩阵与区域连通域面积两组特征,建立特征模板库,利用特征模板库对特钢棒材端面信息码进行识别。结果表明,基于水平双标志点的标记方案能够满足特钢棒材身份标记需求,信息码图像识别方法可有效解决特钢棒材端面信息码的自动识别问题,提高钢厂自动化和智能化水平。
In order to solve the problem of single-root information tracking of special steel bars and improve the level of production intelligence, a special steel bar end information code recognition system based on machine vision was built. According to the environment of the workshop and the characteristics of the end face of special steel bar, the marking scheme of the horizontal double mark points was proposed, and an information code image recognition method was proposed for the marking scheme. The recognition method extracted two sets of features of the character binary image gray level co-occurrence matrix and the regional connected domain area, established a feature template database, and used the feature template database to identify the end face information code of the special steel bar. The results showed that the marking scheme based on the horizontal double mark points can meet the requirements of the identification of special steel bar. The information code image recognition method can effectively solve the problem of the automatic identification of the end face information code of special steel bars and improve the automation and intelligence level of steel mills.
作者
张付祥
赵阳
黄永建
王春梅
黄风山
ZHANG Fu-xiang;ZHAO Yang;HUANG Yong-jian;WANG Chun-mei;HUANG Feng-shan(School of Mechanical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China;Company Office,Shijiazhuang Iron and Steel Co.,Ltd.,Shijiazhuang 050031,Hebei,China;School of Electrical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China)
出处
《中国冶金》
CAS
北大核心
2020年第3期28-34,共7页
China Metallurgy
基金
国家重点研发计划资助项目(2018YFB1308700)
河北省自然科学基金资助项目(E2017208111).
关键词
特钢棒材标记
灰度共生矩阵
区域连通域
特征提取
信息码识别
special steel bar marking
gray level co-occurrence matrix
regional connectivity domain
feature extraction
information code recognition
