Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach
Effective early fire detection is crucial for preventing damage to people and buildings, especially in fire-prone historic structures. However, due to the infrequent occurrence of fire events throughout a building’s lifespan, real-world data for training models are often sparse. In this study, we ap...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-02-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/24/5/1428 |
_version_ | 1797263877375787008 |
---|---|
author | Pascal Vorwerk Jörg Kelleter Steffen Müller Ulrich Krause |
author_facet | Pascal Vorwerk Jörg Kelleter Steffen Müller Ulrich Krause |
author_sort | Pascal Vorwerk |
collection | DOAJ |
description | Effective early fire detection is crucial for preventing damage to people and buildings, especially in fire-prone historic structures. However, due to the infrequent occurrence of fire events throughout a building’s lifespan, real-world data for training models are often sparse. In this study, we applied feature representation transfer and instance transfer in the context of early fire detection using multi-sensor nodes. The goal was to investigate whether training data from a small-scale setup (source domain) can be used to identify various incipient fire scenarios in their early stages within a full-scale test room (target domain). In a first step, we employed Linear Discriminant Analysis (LDA) to create a new feature space solely based on the source domain data and predicted four different fire types (smoldering wood, smoldering cotton, smoldering cable and candle fire) in the target domain with a classification rate up to 69% and a Cohen’s Kappa of 0.58. Notably, lower classification performance was observed for sensor node positions close to the wall in the full-scale test room. In a second experiment, we applied the TrAdaBoost algorithm as a common instance transfer technique to adapt the model to the target domain, assuming that sparse information from the target domain is available. Boosting the data from 1% to 30% was utilized for individual sensor node positions in the target domain to adapt the model to the target domain. We found that additional boosting improved the classification performance (average classification rate of 73% and an average Cohen’s Kappa of 0.63). However, it was noted that excessively boosting the data could lead to overfitting to a specific sensor node position in the target domain, resulting in a reduction in the overall classification performance. |
first_indexed | 2024-04-25T00:19:59Z |
format | Article |
id | doaj.art-583362686af64ee19d50173915a202b6 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-25T00:19:59Z |
publishDate | 2024-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-583362686af64ee19d50173915a202b62024-03-12T16:54:43ZengMDPI AGSensors1424-82202024-02-01245142810.3390/s24051428Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning ApproachPascal Vorwerk0Jörg Kelleter1Steffen Müller2Ulrich Krause3Faculty of Process- and Systems Engineering, Institute of Apparatus and Environmental Technology, Otto von Guericke University of Magdeburg, Universitätsplatz 2, 39106 Magdeburg, GermanyGTE Industrieelektronik GmbH, Helmholtzstr. 21, 38-40, 41747 Viersen, GermanyGTE Industrieelektronik GmbH, Helmholtzstr. 21, 38-40, 41747 Viersen, GermanyFaculty of Process- and Systems Engineering, Institute of Apparatus and Environmental Technology, Otto von Guericke University of Magdeburg, Universitätsplatz 2, 39106 Magdeburg, GermanyEffective early fire detection is crucial for preventing damage to people and buildings, especially in fire-prone historic structures. However, due to the infrequent occurrence of fire events throughout a building’s lifespan, real-world data for training models are often sparse. In this study, we applied feature representation transfer and instance transfer in the context of early fire detection using multi-sensor nodes. The goal was to investigate whether training data from a small-scale setup (source domain) can be used to identify various incipient fire scenarios in their early stages within a full-scale test room (target domain). In a first step, we employed Linear Discriminant Analysis (LDA) to create a new feature space solely based on the source domain data and predicted four different fire types (smoldering wood, smoldering cotton, smoldering cable and candle fire) in the target domain with a classification rate up to 69% and a Cohen’s Kappa of 0.58. Notably, lower classification performance was observed for sensor node positions close to the wall in the full-scale test room. In a second experiment, we applied the TrAdaBoost algorithm as a common instance transfer technique to adapt the model to the target domain, assuming that sparse information from the target domain is available. Boosting the data from 1% to 30% was utilized for individual sensor node positions in the target domain to adapt the model to the target domain. We found that additional boosting improved the classification performance (average classification rate of 73% and an average Cohen’s Kappa of 0.63). However, it was noted that excessively boosting the data could lead to overfitting to a specific sensor node position in the target domain, resulting in a reduction in the overall classification performance.https://www.mdpi.com/1424-8220/24/5/1428multi-sensor nodesearly fire detectiongas sensorstransfer learningelectronic nosefeature fusion |
spellingShingle | Pascal Vorwerk Jörg Kelleter Steffen Müller Ulrich Krause Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach Sensors multi-sensor nodes early fire detection gas sensors transfer learning electronic nose feature fusion |
title | Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach |
title_full | Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach |
title_fullStr | Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach |
title_full_unstemmed | Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach |
title_short | Classification in Early Fire Detection Using Multi-Sensor Nodes—A Transfer Learning Approach |
title_sort | classification in early fire detection using multi sensor nodes a transfer learning approach |
topic | multi-sensor nodes early fire detection gas sensors transfer learning electronic nose feature fusion |
url | https://www.mdpi.com/1424-8220/24/5/1428 |
work_keys_str_mv | AT pascalvorwerk classificationinearlyfiredetectionusingmultisensornodesatransferlearningapproach AT jorgkelleter classificationinearlyfiredetectionusingmultisensornodesatransferlearningapproach AT steffenmuller classificationinearlyfiredetectionusingmultisensornodesatransferlearningapproach AT ulrichkrause classificationinearlyfiredetectionusingmultisensornodesatransferlearningapproach |