Cargando…

Robust Data Augmentation Generative Adversarial Network for Object Detection

Generative adversarial network (GAN)-based data augmentation is used to enhance the performance of object detection models. It comprises two stages: training the GAN generator to learn the distribution of a small target dataset, and sampling data from the trained generator to enhance model performan...

Descripción completa

Detalles Bibliográficos
Autores principales: Lee, Hyungtak, Kang, Seongju, Chung, Kwangsue
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9824481/
https://www.ncbi.nlm.nih.gov/pubmed/36616754
http://dx.doi.org/10.3390/s23010157
_version_ 1784866420652244992
author Lee, Hyungtak
Kang, Seongju
Chung, Kwangsue
author_facet Lee, Hyungtak
Kang, Seongju
Chung, Kwangsue
author_sort Lee, Hyungtak
collection PubMed
description Generative adversarial network (GAN)-based data augmentation is used to enhance the performance of object detection models. It comprises two stages: training the GAN generator to learn the distribution of a small target dataset, and sampling data from the trained generator to enhance model performance. In this paper, we propose a pipelined model, called robust data augmentation GAN (RDAGAN), that aims to augment small datasets used for object detection. First, clean images and a small datasets containing images from various domains are input into the RDAGAN, which then generates images that are similar to those in the input dataset. Thereafter, it divides the image generation task into two networks: an object generation network and image translation network. The object generation network generates images of the objects located within the bounding boxes of the input dataset and the image translation network merges these images with clean images. A quantitative experiment confirmed that the generated images improve the YOLOv5 model’s fire detection performance. A comparative evaluation showed that RDAGAN can maintain the background information of input images and localize the object generation location. Moreover, ablation studies demonstrated that all components and objects included in the RDAGAN play pivotal roles.
format Online
Article
Text
id pubmed-9824481
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-98244812023-01-08 Robust Data Augmentation Generative Adversarial Network for Object Detection Lee, Hyungtak Kang, Seongju Chung, Kwangsue Sensors (Basel) Article Generative adversarial network (GAN)-based data augmentation is used to enhance the performance of object detection models. It comprises two stages: training the GAN generator to learn the distribution of a small target dataset, and sampling data from the trained generator to enhance model performance. In this paper, we propose a pipelined model, called robust data augmentation GAN (RDAGAN), that aims to augment small datasets used for object detection. First, clean images and a small datasets containing images from various domains are input into the RDAGAN, which then generates images that are similar to those in the input dataset. Thereafter, it divides the image generation task into two networks: an object generation network and image translation network. The object generation network generates images of the objects located within the bounding boxes of the input dataset and the image translation network merges these images with clean images. A quantitative experiment confirmed that the generated images improve the YOLOv5 model’s fire detection performance. A comparative evaluation showed that RDAGAN can maintain the background information of input images and localize the object generation location. Moreover, ablation studies demonstrated that all components and objects included in the RDAGAN play pivotal roles. MDPI 2022-12-23 /pmc/articles/PMC9824481/ /pubmed/36616754 http://dx.doi.org/10.3390/s23010157 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/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/).
spellingShingle Article
Lee, Hyungtak
Kang, Seongju
Chung, Kwangsue
Robust Data Augmentation Generative Adversarial Network for Object Detection
title Robust Data Augmentation Generative Adversarial Network for Object Detection
title_full Robust Data Augmentation Generative Adversarial Network for Object Detection
title_fullStr Robust Data Augmentation Generative Adversarial Network for Object Detection
title_full_unstemmed Robust Data Augmentation Generative Adversarial Network for Object Detection
title_short Robust Data Augmentation Generative Adversarial Network for Object Detection
title_sort robust data augmentation generative adversarial network for object detection
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9824481/
https://www.ncbi.nlm.nih.gov/pubmed/36616754
http://dx.doi.org/10.3390/s23010157
work_keys_str_mv AT leehyungtak robustdataaugmentationgenerativeadversarialnetworkforobjectdetection
AT kangseongju robustdataaugmentationgenerativeadversarialnetworkforobjectdetection
AT chungkwangsue robustdataaugmentationgenerativeadversarialnetworkforobjectdetection