Fire Detection Research Articles

Modelling the maximum ceiling temperature with bifurcated plume in tunnel fires

Evaluation of the temperature profile induced by tunnel fire is of great significance for fire detection and structural damage assessment. In this work, a special plume model, namely, the “plume bifurcation” is introduced, and its impact on the temperature distribution is analysed theoretically. Full-scale tunnel fire experiments and numerical simulations are conducted to investigate the plume evolution under various heat release rates and ventilation conditions. Results show that the plume will sperate into two sub-streams under strong ventilation condition, resulting in the multi-dimensional plume movement pattern when rising. The sub-plume impingement point location, quantified by the longitudinal vertical deflection angle θv and horizontal bifurcation angle θh, is correlated with the fire heat release rate and ventilation condition. The plume bifurcation angle is 0 when the dimensionless ventilation velocity V′≤0.33, marking a single-stream plume; and rapidly increase and then slowly decrease with V′ when V′>0.33, representing the sudden occurrence and decay trend of plume bifurcation with the ventilation velocity. Finally, a novel bifurcated plume-based model to estimate the maximum ceiling temperature is proposed. This study reveals the role of multi-dimensional smoke movement in reshaping the temperature field, providing new perspectives for a deeper understanding of the smoke transport behaviour in longitudinally ventilated tunnel fires.

Tiny-Object Detection Based on Optimized YOLO-CSQ for Accurate Drone Detection in Wildfire Scenarios

Wildfires, which are distinguished by their destructive nature and challenging suppression, present a significant threat to ecological environments and socioeconomic systems. In order to address this issue, the development of efficient and accurate fire detection technologies for early warning and timely response is essential. This paper addresses the complexity of forest and mountain fire detection by proposing YOLO-CSQ, a drone-based fire detection method built upon an improved YOLOv8 algorithm. Firstly, we introduce the CBAM attention mechanism, which enhances the model’s multi-scale fire feature extraction capabilities by adaptively adjusting weights in both the channel and spatial dimensions of feature maps, thereby improving detection accuracy. Secondly, we propose an improved ShuffleNetV2 backbone network structure, which significantly reduces the model’s parameter count and computational complexity while maintaining feature extraction capabilities. This results in a more lightweight and efficient model. Thirdly, to address the challenges of varying fire scales and numerous weak emission targets in mountain fires, we propose a Quadrupled-ASFF detection head for weighted feature fusion. This enhances the model’s robustness in detecting targets of different scales. Finally, we introduce the WIoU loss function to replace the traditional CIoU object detection loss function, thereby enhancing the model’s localization accuracy. The experimental results demonstrate that the improved model achieves an mAP@50 of 96.87%, which is superior to the original YOLOV8, YOLOV9, and YOLOV10 by 10.9, 11.66, and 13.33 percentage points, respectively. Moreover, it exhibits significant advantages over other classic algorithms in key evaluation metrics such as precision, recall, and F1 score. These findings validate the effectiveness of the improved model in mountain fire detection scenarios, offering a novel solution for early warning and intelligent monitoring of mountain wildfires.

Correction: A Study of Novel Initial Fire Detection Algorithm Based on Deep Learning Method

Correction: A Study of Novel Initial Fire Detection Algorithm Based on Deep Learning Method

An Improved Fire Detection Algorithm Based on YOLOv8 Integrated with DGIConv, FourBranchAttention and GSIoU

Fire detection is highly important for people's lives and property, and enhancing its accuracy is essential. This study focused on utilizing and improving YOLOv8 to obtain higher detection accuracy for fire detection. Three methods were used. First, the newly designed DGIConv module replaces the original Conv module, thereby decreasing the computational complexity while enhancing the model's performance. Second, to enhance the recognition ability of flame targets, a new attention mechanism named FourBranchAttention was designed, and a comparison was made with other attention mechanisms. The experiments revealed that the newly designed attention mechanism performed best on the mAP50 and mAP50-95 metrics. Finally, to improve the convergence speed and localization ability of the model, the loss function is optimized by adopting better hyperparameters of the TaskAlignedAssigner and employing the newly designed GSIoU as an alternative to the original CIoU. Through ablation experiments, all three improvements improved the detection performance to a certain extent, and the model using the three improvements achieved the best performance. Compared with the baseline, the YOLOv8 model with DGIConv, FourBranchAttention, and the optimized loss function increased the mAP50 by 2.52% and the mAP50-95 by 3.37%. The mAP50 and mAP50-95 had reached 98.46% and 75.26%, respectively. Compared with previous models, such as SSD/YOLOv7, the performance metrics of enhanced YOLOv8 also exhibited significant enhancements, thereby augmenting the accuracy of fire detection. Doi: 10.28991/HIJ-2024-05-03-09 Full Text: PDF

The research and development of an IoT device for early detection of fire and explosion risks and monitoring air quality in computer labs

The research and development of an IoT device for early detection of fire and explosion risks and monitoring air quality in computer labs

Spatial variability in Arctic–boreal fire regimes influenced by environmental and human factors

Wildfire activity in Arctic and boreal regions is rapidly increasing, with severe consequences for climate and human health. Regional long-term variations in fire frequency and intensity characterize fire regimes. The spatial variability in Arctic–boreal fire regimes and their environmental and anthropogenic drivers, however, remain poorly understood. Here we present a fire tracking system to map the sub-daily evolution of all circumpolar Arctic–boreal fires between 2012 and 2023 using 375 m Visible Infrared Imaging Radiometer Suite active fire detections and the resulting dataset of the ignition time, location, size, duration, spread and intensity of individual fires. We use this dataset to classify the Arctic–boreal biomes into seven distinct ‘pyroregions’ with unique climatic and geographic environments. We find that these pyroregions exhibit varying responses to environmental drivers, with boreal North America, eastern Siberia and northern tundra regions showing the highest sensitivity to climate and lightning density. In addition, anthropogenic factors play an important role in influencing fire number and size, interacting with other factors. Understanding the spatial variability of fire regimes and its interconnected drivers in the Arctic–boreal domain is important for improving future predictions of fire activity and identifying areas at risk for extreme events.

Real-Time Monitoring System for Peatland Fire Potential Based on Internet of Things

This research aims to develop a real-time monitoring system for peatland fire potential based on the Internet of Things (IoT) with a focus on early detection of potential peatland fires. The main problem to be solved is the lack of an effective system in the early detection of potential peatland fires, which can cause serious environmental impacts. The method used involves the use of air temperature, air humidity, soil moisture, and fire detection sensors integrated with alarm-based alerts. Data collection is done in real-time to provide a deeper understanding of peatland conditions and potential fire risks. The research results show that the developed system is capable of providing accurate and fast information related to peatland conditions, thus helping to prevent and reduce the impact of peatland fires. With this system, it is expected to increase efficiency in early fire detection and minimize the losses caused by peatland fires.

A hierarchical k-out-of-n optimization model for enhancing reliability of fire alarm systems

Fire alarm systems are crucial for mitigating the risks and damages associated with fires in process industries. These systems heavily depend on the performance of fire detection sensors to provide early detection and warnings, facilitating prompt evacuation and intervention measures. In order to improve the reliability of sensor systems typically requires the use of redundancy allocation techniques, such as a k-out-of-n configuration. In k-out-of-n configuration, the signals from all sensors are directed to a voter, which compares them and transmits a final signal once at least k signals are matched. However, challenges arise due to diverse failure modes with distinct impacts on the system. Integrating sensors through a single-layer k-out-of-n configuration, when there is only one voter, fails to fully highlight the potential advantages of redundancy in sensor systems. This paper addresses the limitations of the conventional configurations by introducing a hierarchical k-out-of-n system in which sensors can be integrated across multiple layers. In the k-out-of-n hierarchical system, sensors are organized into distinct groups, each employing its own k-out-of-n sub-system. Subsequently, the outputs of these sensor groups can be integrated with other groups through their respective voters in higher layers of the hierarchy. Ultimately, at the top layer, all the voters from the preceding layers must be connected to a top voter to transmit a final signal. Furthermore, an optimization model is developed to minimize the system cost while maximizing reliability, considering the best selection of sensors and the system configuration. The proposed optimization model takes into account a multi-sensor system with two competing failure modes and incorporates the probability of sensors' failure modes in the system. The hierarchical k-out-of-n system provides an opportunity to validate sensors by comparing them within their designated groups, allowing for the identification of the location of a failure and the planning of appropriate actions accordingly. The proposed optimization model is applied to a case of a fire detection system in storage warehouse to demonstrate its advantages over the conventional model.

Design and Implementation of a Fire Detection and Extinguishing System Using Dual Axis Mechanics

Fire detection and extinguishing systems play an essential role in anticipating the impact of more severe damage to residential installations, warehouses, and buildings that store valuable goods or important documents. This research develops an automatic fire detection and extinguishing system using the dual-axis rotational principle. The flame channel sensor is used by modifying its default position to be collateral (facing the same plane at the same angle). The dual-axis principle allows panning and tilting movements to yield a broader scope for detecting and extinguishing fires. The Dual Axis built in the research has a panning angle range of 113° and a tilting angle of 45°. The results show good performance with a fire detection accuracy rate of 73-100% and a successful extinguishing rate of 76-96%. The implementation of dual-axis rotational construction in the automatic fire extinguishing system not only improves the detection coverage but also increases the chance ratio of more targeted extinguishing.

ESFD-YOLOv8n: Early Smoke and Fire Detection Method Based on an Improved YOLOv8n Model

Ensuring fire safety is essential to protect life and property, but modern infrastructure and complex settings require advanced fire detection methods. Traditional object detection systems, often reliant on manual feature extraction, may fall short, and while deep learning approaches are powerful, they can be computationally intensive, especially for real-time applications. This paper proposes a novel smoke and fire detection method based on the YOLOv8n model with several key architectural modifications. The standard Complete-IoU (CIoU) box loss function is replaced with the more robust Wise-IoU version 3 (WIoUv3), enhancing predictions through its attention mechanism and dynamic focusing. The model is streamlined by replacing the C2f module with a residual block, enabling targeted feature extraction, accelerating training and inference, and reducing overfitting. Integrating generalized efficient layer aggregation network (GELAN) blocks with C2f modules in the neck of the YOLOv8n model further enhances smoke and fire detection, optimizing gradient paths for efficient learning and high performance. Transfer learning is also applied to enhance robustness. Experiments confirmed the excellent performance of ESFD-YOLOv8n, outperforming the original YOLOv8n by 2%, 2.3%, and 2.7%, with a mean average precision (mAP@0.5) of 79.4%, precision of 80.1%, and recall of 72.7%. Despite its increased complexity, the model outperforms several state-of-the-art algorithms and meets the requirements for real-time fire and smoke detection.

SmokeFireNet: A Lightweight Network for Joint Detection of Forest Fire and Smoke

In recent years, forest fires have been occurring frequently around the globe, affected by extreme weather and dry climate, causing serious economic losses and environmental pollution. In this context, timely detection of forest fire smoke is crucial for realizing real-time early warning of fires. However, fire and smoke from forest fires can spread to cover large areas and may affect distant areas. In this paper, a lightweight joint forest fire and smoke detection network, SmokeFireNet, is proposed, which employs ShuffleNetV2 as the backbone for efficient feature extraction, effectively addressing the computational efficiency challenges of traditional methods. To integrate multi-scale information and enhance the semantic feature extraction capability, a feature pyramid network (FPN) and path aggregation network (PAN) are introduced in this paper. In addition, the FPN network is optimized by a lightweight DySample upsampling operator. The model also incorporates efficient channel attention (ECA), which can pay more attention to the detection of forest fires and smoke regions while suppressing irrelevant features. Finally, by embedding the receptive field block (RFB), the model further improves its ability to understand contextual information and capture detailed features of fire and smoke, thus improving the overall detection accuracy. The experimental results show that SmokeFireNet is better than other mainstream target detection algorithms in terms of average APall of 86.2%, FPS of 114, and GFLOPs of 8.4, and provides effective technical support for forest fire prevention work in terms of average precision, frame rate, and computational complexity. In the future, the SmokeFireNet model is expected to play a greater role in the field of forest fire prevention and make a greater contribution to the protection of forest resources and the ecological environment.

Solar Powered Smart Emergency Light with Fire Detection and Alarm System

Solar Powered Smart Emergency Light with Fire Detection and Alarm System

Domain Adaptation Infrared Forest Fire Detection Method based on YOLOv5 Framework

Domain Adaptation Infrared Forest Fire Detection Method based on YOLOv5 Framework

Experimental Study on Early Fire Smoke Characteristics in a High-Volume Space: A Fire Detection Perspective

Experimental Study on Early Fire Smoke Characteristics in a High-Volume Space: A Fire Detection Perspective

Evolving chimp optimization algorithm using quantum mechanism for engineering applications: a case study on fire detection

Abstract This paper introduces the Quantum Chimp Optimization Algorithm (QU-ChOA), which integrates the Chimp Optimization Algorithm (ChOA) with quantum mechanics principles to enhance optimization capabilities. The study evaluates QU-ChOA across diverse domains, including benchmark tests, the IEEE CEC-06–2019 100-Digit Challenge, real-world optimization problems from IEEE-CEC-2020, and dynamic scenarios from IEEE-CEC-2022. Key findings highlight QU-ChOA’s competitive performance in both unimodal and multimodal functions, achieving an average success rate (SR) of 88.98% across various benchmark functions. QU-ChOA demonstrates robust global search abilities, efficiently finding optimal solutions with an average fitness evaluations (AFEs) of 14 012 and an average calculation duration of 58.22 units in fire detection applications. In benchmark tests, QU-ChOA outperforms traditional algorithms, including achieving a perfect SR of 100% in the IEEE CEC-06–2019 100-Digit Challenge for several functions, underscoring its effectiveness in complex numerical optimization. Real-world applications highlight QU-ChOA’s significant improvements in objective function values for industrial processes, showcasing its versatility and applicability in practical scenarios. The study identifies gaps in existing optimization strategies and positions QU-ChOA as a novel solution to these challenges. It demonstrates QU-ChOA’s numerical advancements, such as a 20% reduction in AFEs compared to traditional methods, illustrating its efficiency and effectiveness across different optimization tasks. These results establish QU-ChOA as a promising tool for addressing intricate optimization problems in diverse fields.

A novel approach based on convolutional neural networks ensemble for fire detection

A novel approach based on convolutional neural networks ensemble for fire detection

IoT-enabled fire detection for sustainable agriculture: A real-time system using flask and embedded technologies

In modern agriculture, the threat of fire poses significant economic and environmental risks. Additionally, traditional fire detection methods are inadequate and poorly integrated with advanced technologies. It is crucial to develop a more efficient and reliable fire detection system. Also, the lack of real-time and accurate monitoring in current systems compromises the resilience and sustainability of farming operations. This article details developing and implementing a real-time fire detection system tailored for smart agriculture. It integrates cutting-edge technologies, including the Internet of Things (IoT), embedded systems, and a Flask-based web application, fortified by cybersecurity measures such as login authentication and secure HTTP protocols. The system's primary aim is to monitor environmental conditions continuously in agricultural fields to detect signs of smoke or flame swiftly, facilitating preventive actions to safeguard crops and infrastructure. The architecture employs sensors distributed across fields to monitor conditions, a Raspberry Pi 3 B+ as the central processing unit for data acquisition and transmission, and a web interface developed using Flask, HTML, and CSS for secure and real-time visualization of detection data. Critical components like the MCP3208 analog-to-digital converter ensure the system's reliability and accuracy. Experimental results confirm the system's efficacy in early fire detection and real-time data visualization via the web browser. Enhanced security features, such as login authentication, protect sensitive information, while regular updates maintain data relevance. This study advances fire prevention and farm preservation efforts, offering a high-performance technological solution for proactive monitoring and quick response to fire risks, thereby supporting food security and sustainable agricultural practices.

A fire detection system based on one-chip computer

A fire detection system based on one-chip computer

Deep Learning Method for Real-Time Fire Detection System for Urban Fire Monitoring and Control

During urban fire incidents, real-time videos and images are vital for emergency responders and decision-makers, facilitating efficient decision-making and resource allocation in smart city fire monitoring systems. However, real-time videos and images require simple and embeddable models in small computer systems with highly accurate fire detection ratios. YOLOv5s has a relatively small model size and fast processing time with limited accuracy. The aim of this study is to propose a method that employs a YOLOv5s network with a squeeze-and-excitation module for image filtering and classification to meet the urgent need for rapid and accurate real-time screening of irrelevant data. In this study, over 3000 internet images were used for crawling and annotating to construct a dataset. Furthermore, the YOLOv5, YOLOv5x and YOLOv5s models were developed to train and test the dataset. Comparative analysis revealed that the proposed YOLOv5s model achieved 98.2% accuracy, 92.5% recall, and 95.4% average accuracy, with a remarkable processing speed of 0.009 s per image and 0.19 s for a 35 frames-per-second video. This surpasses the performance of other models, demonstrating the efficacy of the proposed YOLOv5s for real-time screening and classification in smart city fire monitoring systems.

Prototype smart integrated fire detection based on deep learning YOLO v8 and IoT (internet of things) to improve early fire detection

The high incidence of fires in Indonesia in 2018-2023 is 5,336 fire incidents have caused many deaths and enormous material losses. This system is designed to identify early signs of fire through object detection and sensor technology, which is integrated with the Blynk IoT platform for real-time sensor monitoring and Telegram for instant notifications to users. The waterfall prototype method was designed through observation, system design, program code creation, tool testing, and tool implementation. This research uses Deep Learning YOLOv8 technology and IoT using ESP 32 as a microcontroller. Based on the training datasets, it produces precision=0.95872; recall=0.91; mAP50=0.97; mAP50-95 =0.66. The system uses the integration of a multisensor KY-026 flame sensor, DHT 22 temperature and humidity sensor, and MQ-2 sensors can detect CO, LPG, and smoke gas. All these multisensors can be monitored on Blynk IoT and Telegrambot in real time.