Defects on steel strip surface can majorly cause unhealthy effects, since they make physical and chemical properties mismatched from steel specification. Nowdays , automatic surface detection is used, to find defect on steel strip. These defects appear in major variety of forms and various classes, machine learning methods are basically involved to visual surface inspection for coping with these appearances. In this paper, we present defect detection model to perform defect visualization using convolutional neural network. In the project , the NEU database, which having six kinds of typical surface defects of hot-rolled steel strip, used for improve efficiency of model. The results show that the proposed model can perform defect segmentation in all kinds of defects in database. this process not needed skilled learning with no labeling and small training procedure so it is easy to give required application .also, this defect detection shall improve the productivity and reliability of steel strip's production process.
Keywords- Steel surface defect, Python deep learning, Convolutional Neural Network, Classification.
Software And Hardware
• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL