A lightweight modified YOLOX network using coordinate attention mechanism for PCB surface defect detection

Surface defect detection for the printed circuit board (PCB) is essential in PCB manufacturing. Existing defect detection networks have several problems: low detection efficiency, high memory consumption, and low sensitivity to small defects. To address these issues, we propose a new lightweight dee...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE sensors journal 2022-11, Vol.22 (21), p.1-1
Hauptverfasser:	Xuan, Wang, Jian-She, Gao, Bo-Jie, Hou, Zong-shan, Wang, Hong-wei, Ding, Jie, Wang
Format:	Artikel
Sprache:	eng
Schlagworte:	Circuit boards Computer networks Convolution coordinate attention Datasets Defects Embedded systems Feature extraction Frames per second Lightweight Machine learning modified CSPDarknet Modules Parameters printed circuit board Printed circuits Production Proposals Sensors small defect detection Surface defects Surface treatment Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!