TY - GEN
T1 - FDDNet
T2 - 37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022
AU - Li, Hao
AU - Zhai, Di Hua
AU - Zhao, Shiqi
AU - Liao, Jun
AU - Xia, Yuanqing
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Defect detection is extremely important to improve the quality of PCB production. Although defect detection using traditional methods has achieved good results, a large number of false detections and missed detections cannot be avoided. In response to solve this problems, we propose a fine-grained defect detection network (FDDNet) model to improve the detection performance of PCB defects. This model increases the dimension of spatial context features in PCB defect detection to fuse multi-scale features, which helps the model to deal with more complex scenes. To facilitate the efficiency of feature fusion, we propose an improved channel attention module to enhance the learning efficiency of the network for effective features. To cooperate with the multiplexing of multi-scale feature maps in the backbone network, we propose a module capable of enhancing image recognition to extract pure shallow information. Finally, the experimental results on the PCB defect dataset show that the proposed method can achieve a mAP50 index of 97.32%.
AB - Defect detection is extremely important to improve the quality of PCB production. Although defect detection using traditional methods has achieved good results, a large number of false detections and missed detections cannot be avoided. In response to solve this problems, we propose a fine-grained defect detection network (FDDNet) model to improve the detection performance of PCB defects. This model increases the dimension of spatial context features in PCB defect detection to fuse multi-scale features, which helps the model to deal with more complex scenes. To facilitate the efficiency of feature fusion, we propose an improved channel attention module to enhance the learning efficiency of the network for effective features. To cooperate with the multiplexing of multi-scale feature maps in the backbone network, we propose a module capable of enhancing image recognition to extract pure shallow information. Finally, the experimental results on the PCB defect dataset show that the proposed method can achieve a mAP50 index of 97.32%.
KW - PCB defect detection
KW - deep learning
KW - global spatial context
KW - multi-dimensional attention
UR - http://www.scopus.com/inward/record.url?scp=85147943073&partnerID=8YFLogxK
U2 - 10.1109/YAC57282.2022.10023618
DO - 10.1109/YAC57282.2022.10023618
M3 - Conference contribution
AN - SCOPUS:85147943073
T3 - Proceedings - 2022 37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022
SP - 192
EP - 197
BT - Proceedings - 2022 37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 19 November 2022 through 20 November 2022
ER -