基于2D/3D成像技术的1JC厚板表面质量在线检测系统研发  

Research and development of 2D/3D online surface inspection system for 1JC heavy plate

作　　者：彭铁根[1] 陆东升[2] 何永辉[1] PENG Tiegen;LU Dongsheng;HE Yonghui(Research Institute,Baoshan Iron&Steel Co.,Ltd.,Shanghai 201999,China;Heavy Plate Plant,Baoshan Iron&Steel Co.,Ltd.,Shanghai 200941,China)

机构地区：[1]宝山钢铁股份有限公司中央研究院,上海201999 [2]宝山钢铁股份有限公司厚板厂,上海200941

出　　处：《宝钢技术》2023年第4期60-63,68,共5页Baosteel Technology

摘　　要：针对1JC厚板表面质量在线检测,提出了一种2D/3D复合成像的表面检测系统。应用2D和3D成像系统分别实现厚板上下表面无深度变化及有深度变化缺陷的检测和识别,系统均采用深度学习算法进行分类模型的训练和应用。该技术已在宝钢厚板1JC精整线得到成功验证,现场一年多的应用表明,该技术可以有效实现精整线厚板表面缺陷的在线检测和识别。A surface inspection system based on 2D/3D compound imaging was proposed for the online detection of 1JC heavy plate surface quality.2D and 3D compound imaging systems were used to detect and identify the defects with/without depth change on the upper and lower surfaces of heavy plates respectively,and a deep learning algorithm is used for classification model training and application in all systems.This technology has been verified in the Baosteel 1JC finishing line of heavy plate.The field application for more than one year shows that this technology can effectively realize the online detection and identification of the surface defects of the finished heavy plate.

关 键 词：厚板 表面检测系统 2D/3D复合成像 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]