PCB Defect Detection Method Based on Transformer-YOLO
In order to solve the problem of low accuracy and efficiency in printed circuit board(PCB) defect detection using reference methods, a Transformer-YOLO network detection model is proposed. Firstly, an improved clustering algorithm is used to generate the anchor box suitable for the PCB defect data s...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9978605/ |
| _version_ | 1797980240021028864 |
|---|---|
| author | Wei Chen Zhongtian Huang Qian Mu Yi Sun |
| author_facet | Wei Chen Zhongtian Huang Qian Mu Yi Sun |
| author_sort | Wei Chen |
| collection | DOAJ |
| description | In order to solve the problem of low accuracy and efficiency in printed circuit board(PCB) defect detection using reference methods, a Transformer-YOLO network detection model is proposed. Firstly, an improved clustering algorithm is used to generate the anchor box suitable for the PCB defect data set of this paper. Secondly, abandoning the traditional idea of using convolutional neural network to extract image feature, Swin Transformer is used as the feature extraction network, which can effectively establish the dependency between image features. Finally, to modify the order of the channels in the feature map and enable the network to more effectively focus on the information with greater value, the convolution and attention mechanism module is added to the feature detection network component. Comparing the network model proposed in this paper with Faster R-CNN, SSD, YOLOv3, YOLOv4 and YOLOv5, the experimental results show that the proposed model improves the accuracy by 23.90%, 15.51%, 10.70%, 7.83% and 6.12% respectively, which is better than other most mainstream target detection models and has relatively small volume. |
| first_indexed | 2024-04-11T05:51:23Z |
| format | Article |
| id | doaj.art-a98e6113547048ff979ff3706a204da7 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T05:51:23Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a98e6113547048ff979ff3706a204da72022-12-22T04:42:03ZengIEEEIEEE Access2169-35362022-01-011012948012948910.1109/ACCESS.2022.32282069978605PCB Defect Detection Method Based on Transformer-YOLOWei Chen0https://orcid.org/0000-0002-5382-0542Zhongtian Huang1https://orcid.org/0000-0002-9354-3869Qian Mu2Yi Sun3School of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an, ChinaSchool of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an, ChinaSchool of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an, ChinaSchool of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an, ChinaIn order to solve the problem of low accuracy and efficiency in printed circuit board(PCB) defect detection using reference methods, a Transformer-YOLO network detection model is proposed. Firstly, an improved clustering algorithm is used to generate the anchor box suitable for the PCB defect data set of this paper. Secondly, abandoning the traditional idea of using convolutional neural network to extract image feature, Swin Transformer is used as the feature extraction network, which can effectively establish the dependency between image features. Finally, to modify the order of the channels in the feature map and enable the network to more effectively focus on the information with greater value, the convolution and attention mechanism module is added to the feature detection network component. Comparing the network model proposed in this paper with Faster R-CNN, SSD, YOLOv3, YOLOv4 and YOLOv5, the experimental results show that the proposed model improves the accuracy by 23.90%, 15.51%, 10.70%, 7.83% and 6.12% respectively, which is better than other most mainstream target detection models and has relatively small volume.https://ieeexplore.ieee.org/document/9978605/PCB defectsclustering algorithmconvolutional neural networkSwin Transformerattention mechanism |
| spellingShingle | Wei Chen Zhongtian Huang Qian Mu Yi Sun PCB Defect Detection Method Based on Transformer-YOLO IEEE Access PCB defects clustering algorithm convolutional neural network Swin Transformer attention mechanism |
| title | PCB Defect Detection Method Based on Transformer-YOLO |
| title_full | PCB Defect Detection Method Based on Transformer-YOLO |
| title_fullStr | PCB Defect Detection Method Based on Transformer-YOLO |
| title_full_unstemmed | PCB Defect Detection Method Based on Transformer-YOLO |
| title_short | PCB Defect Detection Method Based on Transformer-YOLO |
| title_sort | pcb defect detection method based on transformer yolo |
| topic | PCB defects clustering algorithm convolutional neural network Swin Transformer attention mechanism |
| url | https://ieeexplore.ieee.org/document/9978605/ |
| work_keys_str_mv | AT weichen pcbdefectdetectionmethodbasedontransformeryolo AT zhongtianhuang pcbdefectdetectionmethodbasedontransformeryolo AT qianmu pcbdefectdetectionmethodbasedontransformeryolo AT yisun pcbdefectdetectionmethodbasedontransformeryolo |