Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach

Fire detection technologies remain a critical component of building automation. With the recent significant advances in computer vision, visual fire detection methods have been developed and integrated into building surveillance systems. Overfitting and accuracy challenges remain in fire detection w...

Full description

Bibliographic Details
Main Authors: Zhaoyi Xu, Yanjie Guo, Joseph Homer Saleh
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9309261/
_version_ 1818639914042916864
author Zhaoyi Xu
Yanjie Guo
Joseph Homer Saleh
author_facet Zhaoyi Xu
Yanjie Guo
Joseph Homer Saleh
author_sort Zhaoyi Xu
collection DOAJ
description Fire detection technologies remain a critical component of building automation. With the recent significant advances in computer vision, visual fire detection methods have been developed and integrated into building surveillance systems. Overfitting and accuracy challenges remain in fire detection when training datasets are limited. In this work, we tackle these challenges by developing a deep convolutional generative adversarial network (DCGAN) for highly accurate visual fire detection when training images are limited. Our model addresses three types of errors in visual fire detection with small training datasets: model overfitting, fire probability overestimation, and fire probability underestimation. The DCGAN includes a generator of fake fire images for self-supervised learning (SSL) and a discriminator for image classification. We designed computational experiments with high-quality datasets to test and validate our model against other supervised learning approaches. We also benchmarked the performance of the DCGAN against a best-in-class deep visual fire detection model. The results show that our model significantly outperforms other fire detection models on all performance metrics when trained with the same small dataset. The results demonstrate that the DCGAN effectively mitigates the three types of error when the training dataset is limited.
first_indexed 2024-12-16T23:02:56Z
format Article
id doaj.art-49cfe9e1d8e64ab5986a12d7e8328c60
institution Directory Open Access Journal
issn 2169-3536
language English
last_indexed 2024-12-16T23:02:56Z
publishDate 2021-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj.art-49cfe9e1d8e64ab5986a12d7e8328c602022-12-21T22:12:39ZengIEEEIEEE Access2169-35362021-01-0193936394610.1109/ACCESS.2020.30477649309261Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network ApproachZhaoyi Xu0https://orcid.org/0000-0002-8498-3483Yanjie Guo1https://orcid.org/0000-0002-1115-7383Joseph Homer Saleh2https://orcid.org/0000-0001-7590-9399School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, USASchool of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, USASchool of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, USAFire detection technologies remain a critical component of building automation. With the recent significant advances in computer vision, visual fire detection methods have been developed and integrated into building surveillance systems. Overfitting and accuracy challenges remain in fire detection when training datasets are limited. In this work, we tackle these challenges by developing a deep convolutional generative adversarial network (DCGAN) for highly accurate visual fire detection when training images are limited. Our model addresses three types of errors in visual fire detection with small training datasets: model overfitting, fire probability overestimation, and fire probability underestimation. The DCGAN includes a generator of fake fire images for self-supervised learning (SSL) and a discriminator for image classification. We designed computational experiments with high-quality datasets to test and validate our model against other supervised learning approaches. We also benchmarked the performance of the DCGAN against a best-in-class deep visual fire detection model. The results show that our model significantly outperforms other fire detection models on all performance metrics when trained with the same small dataset. The results demonstrate that the DCGAN effectively mitigates the three types of error when the training dataset is limited.https://ieeexplore.ieee.org/document/9309261/Deep convolutional generative adversarial networkself-supervised learningvisual fire detection
spellingShingle Zhaoyi Xu
Yanjie Guo
Joseph Homer Saleh
Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
IEEE Access
Deep convolutional generative adversarial network
self-supervised learning
visual fire detection
title Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
title_full Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
title_fullStr Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
title_full_unstemmed Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
title_short Tackling Small Data Challenges in Visual Fire Detection: A Deep Convolutional Generative Adversarial Network Approach
title_sort tackling small data challenges in visual fire detection a deep convolutional generative adversarial network approach
topic Deep convolutional generative adversarial network
self-supervised learning
visual fire detection
url https://ieeexplore.ieee.org/document/9309261/
work_keys_str_mv AT zhaoyixu tacklingsmalldatachallengesinvisualfiredetectionadeepconvolutionalgenerativeadversarialnetworkapproach
AT yanjieguo tacklingsmalldatachallengesinvisualfiredetectionadeepconvolutionalgenerativeadversarialnetworkapproach
AT josephhomersaleh tacklingsmalldatachallengesinvisualfiredetectionadeepconvolutionalgenerativeadversarialnetworkapproach