GAN-based Image Data Augmentation for Improving Object Detection Performance in Industrial Environments (산업 환경의 객체 검출 성능 개선을 위한 GAN 기반 이미지 데이터 증강)

Moon-Ki Back (백문기); Kye Kyung Kim (김계경)

doi:10.29056/jsav.2022.12.25

GAN-based Image Data Augmentation for Improving Object Detection Performance in Industrial Environments

Journal of Software Assessment and Valuation
Abbr : JSAV
2022, 18(2), pp.247-259
DOI : 10.29056/jsav.2022.12.25
Publisher : Korea Software Assessment and Valuation Society
Research Area : Engineering > Computer Science
Received : November 30, 2022
Accepted : December 20, 2022
Published : December 31, 2022

Moon-Ki Back ¹, Kye Kyung Kim ¹

¹한국전자통신연구원

Accredited

ABSTRACT

Object detection is one of the important industrial safety technologies that can automatically provide a worker with alerts to avoid unexpected near misses. However, deep learning-based object detection models require large amounts of training data to achieve higher performance, and data collection and labeling work is laborious and requires human resources. To address these limitations, we propose a GAN-based data augmentation that can supplement the original dataset with more diverse examples. In addition, we present a transformer-based generator network to improve the fidelity of generated data and evaluate the existing object detection model(YOLOv5) trained under different augmentation settings for a comparison study. The evaluation results show that the classification ability of the model trained with 20% augmented data has improved by 0.9% without localization performance losses.

KEYWORDS

Generative Adversarial Networks, Data Augmentation, Object Detection, Deep Learning, Industrial Environment

Citation status

* References for papers published after 2023 are currently being built.

[confproc] Deng Benyang / 2020 / Safety helmet detection method based on YOLO v4 / 2020 16th International Conference on Computational Intelligence and Security (CIS) / IEEE : 155~158

[journal] Nipun D. Nath / 2020 / Deep learning for site safety: Real-time detection of personal protective equipment / Automation in Construction / Elsevier BV 112 : 103085~

[confproc] Congcong Zhao / 2019 / Real-Time Pedestrian Detection Based on Improved YOLO Model / 2019 11th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC) / IEEE 2

[journal] He Xu / 2022 / Vehicle and Pedestrian Detection Algorithm Based on Lightweight YOLOv3-Promote and Semi-Precision Acceleration / IEEE Transactions on Intelligent Transportation Systems / Institute of Electrical and Electronics Engineers (IEEE) 23(10) : 19760~19771

[journal] Alon Halevy / 2009 / The Unreasonable Effectiveness of Data / IEEE Intelligent Systems / Institute of Electrical and Electronics Engineers (IEEE) 24(2) : 8~12

[confproc] Sun C. / 2017 / Revisiting unreasonable effectiveness of data in deep learning era / Proceedings of the IEEE international conference on computer vision

[journal] Muhammad Ahmed / 2021 / Survey and Performance Analysis of Deep Learning Based Object Detection in Challenging Environments / Sensors / MDPI AG 21(15) : 5116~

[web] / ImageNet Large Scale Visual Recognition Challenge (ILSVRC) / http://www.image-net.org/challenges/LSVRC/

[journal] Ian Goodfellow / 2020 / Generative adversarial networks / Communications of the ACM / Association for Computing Machinery (ACM) 63(11) : 139~144

[confproc] Karras T / 2020 / Analyzing and improving the image quality of stylegan / Proceedings of the IEEE/CVF conference on computer vision and pattern recognition : 8110~8119

[journal] Zhao L / 2021 / Improved transformer for high-resolution gans / Advances in Neural Information Processing Systems 34 : 18367~18380

[journal] Maayan Frid-Adar / 2018 / GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification / Neurocomputing / Elsevier BV 321 : 321~331

[confproc] Hojjat Salehinejad / 2018 / Generalization of Deep Neural Networks for Chest Pathology Classification in X-Rays Using Generative Adversarial Networks / 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) / IEEE : 90~994

[journal] Chenglong Wang / 2021 / Lychee Surface Defect Detection Based on Deep Convolutional Neural Networks with GAN-Based Data Augmentation / Agronomy / MDPI AG 11(8) : 1500~

[other] Jiang Y. / 2021 / Transgan: Two transformers can make one strong gan / arXiv preprint arXiv:2102.07074

[journal] Zhao S. / 2020 / Differentiable augmentation for data-efficient gan training / Advances in Neural Information Processing Systems 33 : 7559~7570

[journal] Bokun Liu / 2022 / The Effect of Data Augmentation Methods on Pedestrian Object Detection / Electronics / MDPI AG 11(19) : 3185~

[other] Inoue H. / 2018 / Data augmentation by pairing samples for images classification / arXiv preprint arXiv:1801.02929

[confproc] He K. / 2016 / Deep residual learning for image recognition / Proceedings of the IEEE conference on computer vision and pattern recognition : 770~778

[web] / YOLOv5(Ultralytics) / https://github.com/ultralytics/yolov5

[other] Dosovitskiy A. / 2020 / An image is worth 16x16 words: Transformers for image recognition at scale / arXiv preprint arXiv:2010.11929

[journal] Ali Borji / 2019 / Pros and cons of GAN evaluation measures / Computer Vision and Image Understanding / Elsevier BV 179 : 41~65

[journal] Heusel M. / 2017 / Gans trained by a two time-scale update rule converge to a local nash equilibrium / Advances in neural information processing systems 30

[confproc] Christian Szegedy / 2016 / Rethinking the Inception Architecture for Computer Vision / 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) / IEEE : 2818~2826

[journal] Nobuyuki Otsu / 1979 / A Threshold Selection Method from Gray-Level Histograms / IEEE Transactions on Systems, Man, and Cybernetics / Institute of Electrical and Electronics Engineers (IEEE) 9(1) : 62~66

[confproc] Lin T. Y. / 2014 / Microsoft coco: Common objects in context / European conference on computer vision / Springer : 740~755

KJCKorea
Journal Central

Journal of Software Assessment and Valuation 2023 KCI Impact Factor : 0.35

GAN-based Image Data Augmentation for Improving Object Detection Performance in Industrial Environments

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of Software Assessment and Valuation 2023 KCI Impact Factor : 0.35

GAN-based Image Data Augmentation for Improving Object Detection Performance in Industrial Environments

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (26) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.