Skip to main content

Research Repository

Advanced Search

Msb r‐cnn: A multi‐stage balanced defect detection network

Xu, Zhihua; Lan, Shangwei; Yang, Zhijing; Cao, Jiangzhong; Wu, Zongze; Cheng, Yongqiang

Authors

Zhihua Xu

Shangwei Lan

Zhijing Yang

Jiangzhong Cao

Zongze Wu

Yongqiang Cheng



Abstract

Deep learning networks are applied for defect detection, among which Cascade R‐CNN is a multi‐stage object detection network and is state of the art in terms of accuracy and efficiency. However, it is still a challenge for Cascade R‐CNN to deal with complex and diverse defects, as the widely varied shapes of defects lead to inefficiency for the traditional convolution filter to extract features. Additionally, the imbalance in features, losses and samples cause lower accuracy. To address the above challenges, this paper proposes a multi‐stage balanced R‐CNN (MSB R‐CNN) for defect detection based on Cascade R‐CNN. Firstly, deformable convolution is adopted in different stages of the backbone network to improve its adaptability to the varying shapes of the defect. Then, the features obtained by the backbone network are refined and enhanced by the balanced feature pyramid. To overcome the imbalance of classification and regression loss, the balanced L1 loss is applied at different stages to correct it. Finally, for the sample selection, the interaction of union (IoU) balanced sampler and the online hard example mining (OHEM) sampler are combined at different stages to make the sampling more reasonable, which can bring a better accuracy and convergence effect to the model. The results of our experiments on the DAGM2007 dataset has shown that our network (MSB R‐CNN) can achieve a mean average precision (mAP) of 67.5%, an increase of 1.5% mAP, compared to Cascade R‐CNN.

Citation

Xu, Z., Lan, S., Yang, Z., Cao, J., Wu, Z., & Cheng, Y. (2021). Msb r‐cnn: A multi‐stage balanced defect detection network. Electronics, 10(16), Article 1924. https://doi.org/10.3390/electronics10161924

Journal Article Type Article
Acceptance Date Aug 8, 2021
Online Publication Date Aug 10, 2021
Publication Date Aug 10, 2021
Deposit Date Apr 6, 2022
Publicly Available Date Mar 28, 2024
Journal Electronics (Switzerland)
Electronic ISSN 2079-9292
Publisher MDPI
Peer Reviewed Peer Reviewed
Volume 10
Issue 16
Article Number 1924
DOI https://doi.org/10.3390/electronics10161924
Keywords Multi-stage balanced network; Defect detection; Convolutional neural network
Public URL https://hull-repository.worktribe.com/output/3824894

Files

Published article (3 Mb)
PDF

Publisher Licence URL
http://creativecommons.org/licenses/by/4.0

Copyright Statement
Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).




You might also like



Downloadable Citations