Early Fire Detection System Research Articles

A Multi-Scale Approach to Early Fire Detection in Smart Homes

In recent years, advancements in smart home technologies have underscored the need for the development of early fire and smoke detection systems to enhance safety and security. Traditional fire detection methods relying on thermal or smoke sensors exhibit limitations in terms of response time and environmental adaptability. To address these issues, this paper introduces the multi-scale information transformer–DETR (MITI-DETR) model, which incorporates multi-scale feature extraction and transformer-based attention mechanisms, tailored specifically for fire detection in smart homes. MITI-DETR achieves a precision of 99.00%, a recall of 99.50%, and a mean average precision (mAP) of 99.00% on a custom dataset designed to reflect diverse lighting and spatial conditions in smart homes. Extensive experiments demonstrate that MITI-DETR outperforms state-of-the-art models in terms of these metrics, especially under challenging environmental conditions. This work provides a robust solution for early fire detection in smart homes, combining high accuracy with real-time deployment feasibility.

임베디드 플랫폼을 위한 화재 조기 감지 시스템: 오경보 최소화를 위한 딥러닝 접근 방식

임베디드 플랫폼을 위한 화재 조기 감지 시스템: 오경보 최소화를 위한 딥러닝 접근 방식

Design of smoke detection system using deep learning and sensor fusion with recursive feature elimination cross-validation

The fire safety system is an important component that controls material and immaterial losses. Fire disasters are generally indicated by the appearance of excess smoke and changes in temperature, pressure, and changes in other parameters in the environment. Conventional smoke sensors are limited in reading parameter changes around their environment, making them less effective in early fire detection. This study aims to design a smoke detection system as an early fire detection system, using sensor fusion based on deep learning using the recursive feature elimination method with cross-validation (RFECV) using a random forest classifier used to select optimal parameters from public datasets as the basis for determining the sensor to be used. Based on the RFECV optimal feature, a deep learning algorithm was performed and obtained an accuracy of 0.99, a precision of 0.99, a recall of 1.00, and an F1 score of 0.99, with a latency time of 34.02 μs, which is 71.76% times faster than the original model.

Innovative pre-fire alert smart detection system-based embedded system

Early fire identification is critical to saving lives and property. One viable strategy for improving fire detection systems’ efficacy and precision is the use of Internet of Things (IoT) technology. This article details research into the conception and development of an IoT, ESP32-based, Arduino MEGA-connected smart pre-alert detection system for fires. Because of its flexible nature, the Arduino microcontroller can easily exchange data with various sensors for real-time monitoring and analysis. This study surveys the existing literature to provide insight into many approaches to early fire detection (forecasting), such as sensor fusion methods and the use of the intelligence concept as a pre-predictor. The research highlights the need to set up sophisticated fire monitoring and alarm systems and to develop predictive models based on sensor data. This study illustrates how embedded systems via IoT technology and the ESP32 architecture may improve the efficiency and dependability of very early fire detection systems. This research adds to the expanding corpus of information that may be used to improve fire safety in contemporary settings. The proposal’s sensors include an Arduino MEGA, an ESP32, a MAX6675 module, an MQ-Smoke, and an MQ-Natural Gas for collecting data, which are then processed using Cloud Firebase and a sleek mobile interface that can be reached overseas at any time.

Early Fire Detection System by Using Automatic Synthetic Dataset Generation Model Based on Digital Twins

Fire is amorphous and occurs differently depending on the space, environment, and material of the fire. In particular, the early detection of fires is a very important task in preventing large-scale accidents; however, there are currently almost no learnable early fire datasets for machine learning. This paper proposes an early fire detection system optimized for certain spaces using a digital-twin-based automatic fire learning data generation model for each space. The proposed method first automatically generates realistic particle-simulation-based synthetic fire data on an RGB-D image matched to the view angle of a monitoring camera to build a digital twin environment of the real space. In other words, our method generates synthetic fire data according to various fire situations in each specific space and then performs transfer learning using a state-of-the-art detection model with these datasets and distributes them to AIoT devices in the real space. Synthetic fire data generation optimized for a space can increase the accuracy and reduce the false detection rate of existing fire detection models that are not adaptive to space.

Implementasi NodeMCU dan Blynk Dalam Sistem Pendeteksi Kebakaran Berbasis IoT

Fires can cause material losses and human lives. Therefore, an effective fire early detection system is needed to prevent greater losses. In this research, ESP8266 is used as a microcontroller to connect fire detection sensors such as KY-026 Fire Sensor and MQ-2 Smoke Sensor. This early fire detection system is also equipped with a buzzer as an alarm and NodeMCU as a platform to connect the system to the Internet network and Blynk as a medium to provide real-time fire notifications to users. Tests were conducted by comparing the response time and accuracy of fire detection at different distances. The test results show that the fire early detection system built has an average response time of 2 seconds and a detection accuracy rate of 95%.

Features of the coefficient of variation of parameters of the gas environment in fire in the premises

The object of the study is the selective coefficient of variation of dangerous parameters of the gas environment during the ignition of materials. The measure of the sample coefficient of variation of an arbitrary hazardous parameter of the gas environment observed at an arbitrary time interval is substantiated. The representativeness error of the measure of the sampling coefficient of variation, which depends on the value of the measure and the sample size, was determined. The measure allows you to numerically determine its value for an arbitrary observation interval. The difference in the measure at the intervals corresponding to the reliable absence and occurrence of ignition allows to detect the occurrence of ignition of the material. According to the results of laboratory studies, the measures of the sample coefficient of variation for carbon monoxide concentration, smoke density, and temperature of the gas medium in the laboratory chamber at intervals of absence and appearance of ignition of alcohol, paper, wood, and textiles were determined. It was established that the dangerous parameters of the gas environment at the intervals of absence and presence of ignition are characterized by different values of the increase in the measure of the sample coefficient of variation. For example, it is determined that the ignition of alcohol causes the maximum increase in the measure for carbon monoxide concentration from 0.135 to 0.441, for smoke density from 0.629 to 0.805, and for temperature from 0.001 to 0.115. When paper catches fire, the measure for carbon monoxide concentration and temperature increases from 0.0026 to 0.140 and from 0.0019 to 0.05, respectively. When burning wood, the measure for carbon monoxide concentration and temperature increases from 0.0072 to 0.177 and from 0.0067 to 0.016, respectively. The obtained results, provided that the hazardous parameters of the gas environment in the premises are measured and the sample coefficient of variation is calculated in practice, make it possible to use them in the creation of early fire detection systems

Application of Smart and Early Fire Detection Systems in Hospitals: Insights from Iraq

Application of Smart and Early Fire Detection Systems in Hospitals: Insights from Iraq

Wireless Sensor Network Based Early Fire Detection System

Every year millions of hectares of forest are devastated by fire. Forest fires may occur due to Natural causes or Man-made causes. Natural causes include lightning which set trees on fire, High atmospheric temperatures and dryness (low humidity). Man-made causes include naked flame, cigarette or bidi, electric spark or any source of ignition that comes into contact with inflammable material. Reason could be any, but important is forest fire causes huge damage to forest and nature. Infact, fires on large scales cause air pollution, mar quality of stream water, threaten biodiversity and spoil the aesthetics of an area. Forest fire causes imbalances in nature and endangers biodiversity by reducing faunal and floral wealth. Therefore, there is a need to develop methods for timely detection of forest fire so that damage is minimum. Existed methods for forest monitoring and fire detection are traditional and based on human observation, satellite imaging, use of digital cameras. There are several drawbacks associated with them such as inefficiency, power consumption, latency, accuracy and implementation costs. Therefore, to address these problems the deployment of automatic fire detection system is necessary to allow early and fast fire extinction. In this paper we present an early fire detection system based on Wireless Sensor Network.

Distance-Based Analysis of Early Fire Indicators on a New Indoor Laboratory Dataset with Distributed Multi-Sensor Nodes

This work analyzes a new indoor laboratory dataset looking at early fire indicators in controlled and realistic experiments representing different incipient fire scenarios. The experiments were performed within the constraints of an indoor laboratory setting using multiple distributed sensor nodes in different room positions. Each sensor node collected data of particulate matter (PM), volatile organic compounds (VOCs), carbon monoxide (CO), carbon dioxide (CO2), hydrogen (H2), ultraviolet radiation (UV), air temperature, and humidity in terms of a multivariate time series. These data hold immense value for researchers within the machine learning and data science communities who are keen to explore innovative and advanced statistical and machine learning techniques. They serve as a valuable resource for the development of early fire detection systems. The analysis of the collected data was carried out depending on the Manhattan distance between the fire source and the sensor node. We found that especially larger particles (>0.5 μm) and VOCs show a significant dependency with respect to the intensity as a function of the Manhattan distance to the source. Moreover, we observed differences in the propagation behavior of VOCs, PM, and CO, which are particularly relevant in incipient fire scenarios due to the presence of strand propagation effects.

Deep learning and stereo vision based detection of post-earthquake fire geolocation for smart cities within the scope of disaster management: İstanbul case

Cities, which are getting crowded day by day, face a number of issues, including increased planning, inadequate infrastructure, heavy traffic and security. Due to such issues, the cities are more susceptible to natural disasters. Although many preparations are made against natural disasters in urban life, it is evident that adequate measures are not taken against secondary disasters. One of the most destructive and frequent secondary disasters is fires after earthquakes. In this study, an early fire detection system is proposed in order to minimize the losses caused by fires in the potentially chaotic environment that may occur after earthquakes in cities. This system consists of a structure that detects fire with convolutional neural network (CNN) based You Only Look Once (YOLO) model, determines the geolocation of the fire with stereo vision/epipolar geometry and provides information to the disaster management center via wireless sensor network (WSN). Experimental test findings conducted in İstanbul verify that the proposed system would be useful for post-earthquake fire detection with low cost, high reliability and high accuracy.

A Design of Early Fire Detection System In Tobacco Oven Based on Internet of Things (Case Study: East Landah Praya Village)

A tobacco oven is a building measuring 4.5 x 4.5 meters that are used to bake freshly harvested tobacco leaves and can accommodate up to 5-7 shelves. The oven building is equipped with 2 bottom vents and 2 top vents that function as temperature and humidity regulators by opening or closing the vents. Tobacco oven techniques in Lombok itself can not be said to be safe, it is proven that fires in tobacco ovens often occur lately. One of the causes of tobacco oven fires is due to the presence of tobacco leaves that fall into the furnace so that the fire quickly grabs the tobacco in a semi-dry state, in addition, due to the lack of use of safety equipment above the ignition furnace, the fallen dried tobacco leaves can immediately burn. Based on these problems, research entitled "Design of Fire Early Detection System in Tobacco Oven Based on Internet Of Things was made (Case Study: East Landah Praya Village)". The system built shows the ability of the infrared fire sensor to detect the presence of fire up to 77.7% at a large fire intensity, the MQ-135 smoke sensor is set at more than 3 ppm as an indication of a fire and the DHT22 temperature sensor shows 1.5% error after comparison with a thermometer. in a tobacco oven.

An AI-Based Early Fire Detection System Utilizing HD Cameras and Real-Time Image Analysis

Wildfires pose a significant threat to human lives, property, and the environment. Rapid response during a fire's early stages is critical to minimizing damage and danger. Traditional wildfire detection methods often rely on reports from bystanders, leading to delays in response times and the possibility of fires growing out of control. In this paper, ask the question: “Can AI object detection improve wildfire detection and response times?”. We present an innovative early fire detection system that leverages state-of-the-art hardware, artificial intelligence (AI)-powered object detection, and seamless integration with emergency services to significantly improve wildfire detection and response times. Our system employs high-definition panoramic cameras, solar-powered energy sources, and a sophisticated communication infrastructure to monitor vast landscapes in real-time. The AI model at the core of the system analyzes images captured by the cameras every 60 seconds, identifying early smoke patterns indicative of fires, and promptly notifying the fire department. We detail the system architecture, AI model framework, training process, and results obtained during testing and validation. The system demonstrates its effectiveness in detecting and reporting fires, reducing response times, and improving emergency services coordination. We have demonstrated that AI object detection can be an invaluable tool in the ongoing battle against wildfires, ultimately saving lives, property, and the environment.

Development of a mid‐infrared early fire detection system based on dual optical path nonresonant photoacoustic cell

AbstractA mid‐infrared carbon dioxide (CO2) sensor system was developed for early fire detection based on photoacoustic spectroscopy. A miniaturized nonresonant photoacoustic gas cell was optimized by means of finite element method (FEM). A cylindrical cell was optimized with a radius of 12.7 mm and a length of 32 mm. An approximately proportional relationship between the optical path length and the amplitude of the photoacoustic signal was observed by theoretical analysis. The optimal modulation frequency was determined to be 44 Hz. A dual optical path was adopted to gain a low limit of detection (LoD) and a high sensitivity. The amplitude of the photoacoustic signal was increased by 1.8 times compared to single optical path. The LoD of the sensor system was measured to be 4.1 ppmv at an averaging time of 84 s.

Forest Fire Detection and Notification Method Based on AI and IoT Approaches

There is a high risk of bushfire in spring and autumn, when the air is dry. Do not bring any flammable substances, such as matches or cigarettes. Cooking or wood fires are permitted only in designated areas. These are some of the regulations that are enforced when hiking or going to a vegetated forest. However, humans tend to disobey or disregard guidelines and the law. Therefore, to preemptively stop people from accidentally starting a fire, we created a technique that will allow early fire detection and classification to ensure the utmost safety of the living things in the forest. Some relevant studies on forest fire detection have been conducted in the past few years. However, there are still insufficient studies on early fire detection and notification systems for monitoring fire disasters in real time using advanced approaches. Therefore, we came up with a solution using the convergence of the Internet of Things (IoT) and You Only Look Once Version 5 (YOLOv5). The experimental results show that IoT devices were able to validate some of the falsely detected fires or undetected fires that YOLOv5 reported. This report is recorded and sent to the fire department for further verification and validation. Finally, we compared the performance of our method with those of recently reported fire detection approaches employing widely used performance matrices to test the achieved fire classification results.

Hybrid Feature Fusion-Based High-Sensitivity Fire Detection and Early Warning for Intelligent Building Systems

High-sensitivity early fire detection is an essential prerequisite to intelligent building safety. However, due to the small changes and erratic fluctuations in environmental parameters in the initial combustion phase, it is always a challenging task. To address this challenge, this paper proposes a hybrid feature fusion-based high-sensitivity early fire detection and warning method for in-building environments. More specifically, the temperature, smoke concentration, and carbon monoxide concentration were first selected as the main distinguishing attributes to indicate an in-building fire. Secondly, the propagation neural network (BPNN) and the least squares support vector machine (LSSVM) were employed to achieve the hybrid feature fusion. In addition, the genetic algorithm (GA) and particle swarm optimization (PSO) were also introduced to optimize the BPNN and the LSSVM, respectively. After that, the outputs of the GA-BPNN and the PSO-LSSVM were fused to make a final decision by means of the D-S evidence theory, achieving a highly sensitive and reliable early fire detection and warning system. Finally, an early fire warning system was developed, and the experimental results show that the proposed method can effectively detect an early fire with an accuracy of more than 96% for different types and regions of fire, including polyurethane foam fire, alcohol fire, beech wood smolder, and cotton woven fabric smolder.

Multisensor Information Fusion for Condition Based Environment Monitoring

Destructive wildfires are becoming an annual event, similar to climate change, resulting in catastrophes that wreak havoc on both humans and the environment. The result, however, is disastrous, causing irreversible damage to the ecosystem. The location of the incident and the hotspot can sometimes have an impact on early fire detection systems. With the advancement of intelligent sensor-based control technologies, the multi-sensor data fusion technique integrates data from multiple sensor nodes. The primary objective to avoid wildfire is to identify the exact location of wildfire occurrence, allowing fire units to respond as soon as possible. Thus to predict the occurrence of fire in forests, a fast and effective intelligent control system is proposed. The proposed algorithm with decision tree classification determines whether fire detection parameters are in the acceptable range and further utilizes a fuzzy-based optimization to optimize the complex environment. The experimental results of the proposed model have a detection rate of 98.3. Thus, providing real-time monitoring of certain environmental variables for continuous situational awareness and instant responsiveness.

The Design of Microcontroller Based Early Warning Fire Detection System for Home Monitoring

Fire is a type of disaster that can occur anytime and anywhere as a result of any accidental or intentional causes. Without exception, houses are also very vulnerable to fire. To anticipate the catastrophic effects of fire that can destroy houses, advanced technology, such as the Internet of Things (IoT) can be utilized to detect the smoke and fire. This study aims to design an early warning fire detection system for home monitoring using smoke detection sensors based on Arduino microcontroller together with NodeMCU ESP8266. This early warning fire detection system is expected to function by notifying homeowners when detecting the presence of smoke in their homes. With the aid of this detection system, the issue of potential damage, death, or material loss caused by fire can be significantly reduced. The results and testing of the designed system will be discussed in the paper.

Smart Video-Based Threat Analysis and Detection using CNN

In recent years, more and more video surveillance devices like drones, CCTV's have been deployed due to an increase in demands related to public security and smart cities. There is a need to overcome the existing drawbacks of post-investigation techniques of video surveillance systems by providing a pre-alert generation system. The video surveillance system has become an important part of the security and protection of modern cities. So we are going to focus on video surveillance by giving video contents containing early fire events detection, suspicious activities and smart parking systems, and crowd estimation. Smart monitoring cameras equipped with intelligent video analytics techniques can monitor and pre-alert systems by capturing suspicious activity and events. Our work is based on deep learning techniques for video analysis with better performance and event detection with the advantages of alert generation.

Early Fire Detection: A New Indoor Laboratory Dataset and Data Distribution Analysis

Fire alarm systems are typically equipped with various sensors such as heat, smoke, and gas detectors. These provide fire alerts and notifications of emergency exits when a fire has been detected. However, such systems do not give early warning in order to allow appropriate action to be taken when an alarm is first triggered, as the fire may have already caused severe damage. This paper analyzes a new dataset gathered from controlled realistic fire experiments conducted in an indoor laboratory environment. The experiments were conducted in a controlled manner by triggering the source of fire using electrical devices and charcoal on paperboard, cardboard or clothing. Important data such as humidity, temperature, MQ139, Total Volatile Organic Compounds (TVOC) and eCO2 were collected using sensor devices. These datasets will be extremely valuable to researchers in the machine learning and data science communities interested in pursuing novel advanced statistical and machine learning techniques and methods for developing early fire detection systems. The analysis of the collected data demonstrates the possibility of using eCO2 and TVOC reading levels for early detection of smoldering fires. The experimental setup was based on Low-Power Wireless Area Networks (LPWAN), which can be used to reliably deliver fire-related data over long ranges without depending on the status of a cellular or WiFi Network.