A semi-supervised convolutional neural network-based method for steel surface defect recognition

•Labeling large-scale data for steel surface defect recognition is costly and hard.•A semi-supervised learning method is proposed with limited labeled data.•This method has better performances with 17.53% improvement.•The proposed method is successfully applied into a real-world case. Automatic defe...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Robotics and computer-integrated manufacturing 2020-02, Vol.61, p.101825, Article 101825
Hauptverfasser:	Gao, Yiping, Gao, Liang, Li, Xinyu, Yan, Xuguo
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial neural networks Convolutional Neural Network Defects Iron and steel making Neural networks Pseudo-Label Recognition Semi-supervised learning Steel industry Steel making Steel production Steel surface defect recognition Surface defects Training
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!