DBFN: Double Branch Fusion Network for Vital Components and Defect Detection of Transmission Line. Issue 4 (19th February 2023)
- Record Type:
- Journal Article
- Title:
- DBFN: Double Branch Fusion Network for Vital Components and Defect Detection of Transmission Line. Issue 4 (19th February 2023)
- Main Title:
- DBFN: Double Branch Fusion Network for Vital Components and Defect Detection of Transmission Line
- Authors:
- Xu, Wenxiao
Chu, Shengguang
Yang, Shanshan - Abstract:
- Abstract: A double branch fusion network is proposed based on unmanned aerial vehicle (UAV) inspection images to increase the detection accuracy of vital components and defects in transmission lines. The backbone feature extraction network comprises a combination of a convolutional neural network (CNN) and a Transformer network. To be specific, the CNN should extract local information, and the Transformer network is responsible for the extraction of global information. Besides, global information and local information have semantic differences, while resulting in feature aliasing after fusion. To solve this problem, a multiscale convolution module and a multiscale pooling module are proposed to solve semantic differences and feature aliasing through the interaction between two types of information. In general, the enhanced feature extraction network comprises a residual‐like convolution module, which can reduce the loss of detailed information (e.g., edge contours) and further extract high‐level semantic information from the deep network. Besides, it performs feature fusion in multiple regions in the enhanced feature extraction network, such that the multi‐scale adaptability of the neural network is effectively enhanced. Last, the fused feature information at different scales is decoded, and the final detection results are yielded. Abstract : To improve the detection accuracy of vital components and defects in transmission lines. Based on unmanned aerial vehicle inspectionAbstract: A double branch fusion network is proposed based on unmanned aerial vehicle (UAV) inspection images to increase the detection accuracy of vital components and defects in transmission lines. The backbone feature extraction network comprises a combination of a convolutional neural network (CNN) and a Transformer network. To be specific, the CNN should extract local information, and the Transformer network is responsible for the extraction of global information. Besides, global information and local information have semantic differences, while resulting in feature aliasing after fusion. To solve this problem, a multiscale convolution module and a multiscale pooling module are proposed to solve semantic differences and feature aliasing through the interaction between two types of information. In general, the enhanced feature extraction network comprises a residual‐like convolution module, which can reduce the loss of detailed information (e.g., edge contours) and further extract high‐level semantic information from the deep network. Besides, it performs feature fusion in multiple regions in the enhanced feature extraction network, such that the multi‐scale adaptability of the neural network is effectively enhanced. Last, the fused feature information at different scales is decoded, and the final detection results are yielded. Abstract : To improve the detection accuracy of vital components and defects in transmission lines. Based on unmanned aerial vehicle inspection image, a double branch fusion network is proposed, which is composed of backbone feature extraction network, enhanced feature extraction network, and decoder. The backbone feature extraction is composed of convolutional neural network and Transformer. … (more)
- Is Part Of:
- Advanced theory and simulations. Volume 6:Issue 4(2023)
- Journal:
- Advanced theory and simulations
- Issue:
- Volume 6:Issue 4(2023)
- Issue Display:
- Volume 6, Issue 4 (2023)
- Year:
- 2023
- Volume:
- 6
- Issue:
- 4
- Issue Sort Value:
- 2023-0006-0004-0000
- Page Start:
- n/a
- Page End:
- n/a
- Publication Date:
- 2023-02-19
- Subjects:
- convolutional neural network -- target detection -- transformer -- unmanned aerial vehicle
Science -- Simulation methods -- Periodicals
Science -- Methodology -- Periodicals
Engineering -- Simulation methods -- Periodicals
Engineering -- Methodology -- Periodicals
507.21 - Journal URLs:
- http://onlinelibrary.wiley.com/ ↗
- DOI:
- 10.1002/adts.202200691 ↗
- Languages:
- English
- ISSNs:
- 2513-0390
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 0696.935575
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 26949.xml