Fire Monitoring Research Articles

A Novel Method for Analyzing Infrared Images Taken by Unmanned Aerial Vehicles for Forest Fire Monitoring

A Novel Method for Analyzing Infrared Images Taken by Unmanned Aerial Vehicles for Forest Fire Monitoring

UNMANNED AERIAL VEHICLES IN OPERATIONAL ACTIVITIES EXAMPLE OF SELECTED VOLUNTEER FIRE BRIGADE UNITS

This article presents the results of research concerning the use of Unmanned Aerial Vehicles(UAVs) in operations in selected Voluntary Fire Brigade (VFB) units. In addition to literatureresearch, a survey method was used. A SWOT analysis was also carried out based on the answersprovided in the questionnaires. The provided information shows that VFBs most often use UAVsfor searching for missing persons, monitoring mass events, tracking illegal rubbish dumps, aswell as during firefighting operations including monitoring of large area fires. Also noted weresuch cases of UAV usage as smoke surveys from chimneys, monitoring of internal fires, locatingwild boar herds during African swine fever (ASF) or initial assessment of the health of an injuredperson. As part of the SWOT analysis, 12 strengths and weaknesses of the UAV were identified,as well as 5 opportunities and threats related to their implementation and application. The articleends with conclusions and recommendations for further research and implementation related toUAVs. This may serve to assure further development of this technology and give an overview ofany pros and cons of its implementation in operational activities in the general security and safetysector, including rescue units.

A COMPARATIVE STUDY ON FOREST WILDFIRE RISK MITIGATION AND MANAGEMENT IN INDONESIA AND UNITED STATES (CASE STUDY: JAMBI AND CALIFORNIA)

It has been known globally that Indonesia has the largest forest area in the world and is often called as World’s lungs. However, wildfire disasters in Indonesia’s forests are happening every year in various locations. This paper will study the cases by focusing on a case study in Jambi Province and compare it with similar issues that happened in California, United States. The paper shows that various human activities are the main factor in how the accident has happened over the last couple of years. Non-environmental-friendly Activities such as illegal logging and fire land clearing for palm oil plantations have been conducted by private institutions and causing long-lasting impacts on local communities. The poor knowledge of local people on how to prevent and handle wildfire forests also has become an issue that needs to be solved as soon as possible. Compared with what happened in California, the main causes are different considering the fires are caused by high temperatures and climate change’s impact. Although the local community in California barely participates in fire management, the mitigation and monitoring infrastructure is much more advanced compared to Indonesia’s.

Forest Fire Monitoring Method Based on UAV Visual and Infrared Image Fusion

Forest fires have become a significant global threat, with many negative impacts on human habitats and forest ecosystems. This study proposed a forest fire identification method by fusing visual and infrared images, addressing the high false alarm and missed alarm rates of forest fire monitoring using single spectral imagery. A dataset suitable for image fusion was created using UAV aerial photography. An improved image fusion network model, the FF-Net, incorporating an attention mechanism, was proposed. The YOLOv5 network was used for target detection, and the results showed that using fused images achieved a higher accuracy, with a false alarm rate of 0.49% and a missed alarm rate of 0.21%. As such, using fused images has greater significance for the early warning of forest fires.

Decolonizing wildfire risk management: indigenous responses to fire criminalization policies and increasingly flammable forest landscapes in Lomerío, Bolivia

Drawing on decolonial thought, this article provides a perspective on local indigenous knowledge and governance systems as a resource for informing wildfire risk policy approaches and collaborative environmental security. In 2019, the Indigenous Territory of Lomerío in Bolivia was heavily affected by wildfires, due to a combination of fires that penetrated the territory from outside and others that spread from inside. As result, the Bolivian Forest Management Agency (ABT) started threatening indigenous people with criminal action for using fire in their livelihood practices. In response, in 2020 and with the support of several institutions, the Union of Indigenous Communities of Lomerío (CICOL) initiated a series of activities to ensure local control of wildfire risk management in the territory. These include a written burning protocol, a fire monitoring programme, water basin and forest conservation policies, participatory research conducted by indigenous researchers about the use of fire in Lomerío and cultural revitalization strategies. The article presents the results of these different strategies and their contributions to creating awareness of appropriate regulations for wildfire risk management by national authorities from the perspective of the Monkoxɨ indigenous people.

Development of Cotton Picker Fire Monitoring System Based on GA-BP Algorithm.

Due to the characteristics of the cotton picker working in the field and the physical characteristics of cotton, it is easy to burn during the operation, and it is difficult to be detected, monitored, and alarmed. In this study, a fire monitoring system of cotton pickers based on GA optimized BP neural network model was designed. By integrating the monitoring data of SHT21 temperature and humidity sensors and CO concentration monitoring sensors, the fire situation was predicted, and an industrial control host computer system was developed to monitor the CO gas concentration in real time and display it on the vehicle terminal. The BP neural network was optimized by using the GA genetic algorithm as the learning algorithm, and the data collected by the gas sensor were processed by the optimized network, which effectively improved the data accuracy of CO concentration during fires. In this system, the CO concentration in the cotton box of the cotton picker was validated, and the measured value of sensor was compared with the actual value, which verified the effectiveness of the optimized BP neural network model with GA. The experimental verification showed that the system monitoring error rate was 3.44%, the accurate early warning rate was over 96.5%, and the false alarm rate and the missed alarm rate were less than 3%. In this study, the fire of cotton pickers can be monitored in real time and an early warning can be made in time, and a new method was provided for accurate monitoring of fire in the field operation of cotton pickers.

The Safety Monitoring System of University Laboratory Based on Cloud Platform

The unified laboratory fire monitoring system aims to centrally monitor laboratory environmental data across the province or even the country, perceive real-time changes in the laboratory's internal environment, accurately assess fire risks, and take timely disaster relief measures. In addition, once a fire occurs in the laboratory, an alarm message can be issued in a timely manner, which can better extinguish the fire in a timely manner to reduce losses or automatically take fire extinguishing measures to avoid the occurrence of a fire. For this purpose, this article designs a cloud platform university laboratory security monitoring system based on ZigBee. This system can monitor the temperature, humidity, and smoke concentration in laboratory rooms of different colleges and universities, and conduct safety monitoring of the rooms. When the monitoring values exceed the safety range, real-time alarms will be given. This system consists of four parts: terminal data acquisition module, ZigBee coordinator, ESP8266 wireless gateway communication module, OneNET cloud platform, and third-party web application. The system collects monitoring data in the laboratory through various sensors and wirelessly transmits the data to the cloud through ZigBee networking. The cloud then transmits the data to the web, thereby achieving real-time monitoring and alerting of the security situation in each laboratory. This article provides a human-machine interaction page on the local web end and a cloud platform monitoring interface. The effectiveness of the system is verified by the measured data validation.

Fire Impacts on Water Resources: A Remote Sensing Methodological Proposal for the Brazilian Cerrado

Fire events are increasing in frequency, duration, and severity worldwide. The combination of ash and uncovered land might increase the transportation of pollutants into the streams, potentially affecting the water supply systems. The intensifying fires in Brazil’s Cerrado biome, responsible for 70% of the country’s water supply, give rise to profound ecological, climatic, and socio-economic concerns that require urgent and effective mitigation strategies. However, little attention has been paid to the consequences of fire events on water resources in the region. In this study, the Fire Impact on Water Resources Index (FIWRI) is proposed and applied in six different water supply watersheds to analyse fire behaviour from 2003 to 2020 and its potential impact on inland water bodies. This is the first remote-sensing-based index for fire impact on water resources developed for the Brazilian territory, to support water management on a watershed scale and uses variables such as terrain slope, river proximity, and vegetation to classify fire events as having a low to high potential to contaminate water bodies. We observed that all six water supply watersheds suffered frequent fire events, with different FIWRI proportions, which ranged from High to Low FIWRI. The proposed index could be used in real-time fire monitoring alert systems in order to support water supply management.

Geostationary active fire products validation: GOES-17 ABI, GOES-16 ABI, and Himawari AHI

ABSTRACT The current generation of geostationary Earth-observing satellites provide spectral bandpass, spatial resolution and imaging frequency characteristics well suited to near-continuous active fire detection and monitoring. The earliest of these systems-SEVIRI on-board EUMETSAT’s MSG series-has operated since 2004, and more recently the capability has been expanded globally with the ABI on-board NOAA’s GOES-16 and GOES-17 satellites, and the AHI on-board JMA’s Himawari-8 and Himawari-9. At present, the NOAA and EUMETSAT operational geostationary active fire products are available based on two different algorithms: the Fire Detection and Characterization (FDC) product operating with data from GOES-16 and −17, and FRP-PIXEL active fire products from GOES, Himawari and MSG. We have conducted a comprehensive accuracy assessment of these geostationary fire products across two seasons (1 January–31 March 2020 and 1 July–30 September 2020), based on comparison to Landsat active fire detections made simultaneously (±5 minutes of geostationary overpass time) with the geostationary data. Compared to Landsat we find (i) low false alarm rates, ranging between 4%–7% (FDC) and 2%-6% (FRP-PIXEL)- depending on the season and hemispheric-disk for high confidence pixels, (ii) a reduction in this false alarm rate for FDC due to algorithm changes made since our prior (2018) validation effort (48% false alarms in summer 2018 compared to 4% in summer 2020 for high confidence pixels), and (iii) comparable active fire pixel detection rates for the FDC product (high confidence fire pixel classes only) and the matching FRP-PIXEL product (all fire pixel confidence classes). Overall, the performance of these geostationary products is shown to be strong and complementary in that the FRP-PIXEL product has fewer false alarms but a lower detection rate, whereas the FDC product detects more fire pixels but with a much higher false alarm rate.

IoT and health monitoring wearable devices as enabling technologies for sustainable enhancement of life quality in smart environments

The Internet of Things (IoT) technology with wearable devices provides a promising solution that enables con-tinuous monitoring of health parameters. Non-invasive sensors enable remote health monitoring and improve many aspects of life quality. The study explores advances in IoT and Wearable Devices through a conceptualization of health monitoring observer networks. The paper proposes an architecture inspired by a formal description of human perception, which has been successfully employed in environmental monitoring and can be adapted to healthcare for better integration and comprehension of patient health data. The proposed architecture involves a three-layer ap-proach Sensors, Integration and Association and Application and Diagnostic, similar to the architecture used in forest fire monitoring. The Sensors layer describes state-of-the-art cutting-edge equipment and related sensors for gathering information. The second layer, Integration and Association propose an ontology and a potential solution for integrat-ing sensor information from the device to the cloud. Furthermore, the third layer, Application and Diagnostic gives a review of used machine learning algorithms for detecting and diagnosing various diseases. Finally, a case study of elderly health monitoring using wearable devices is presented, discussing the methodology for economic value calcu-lation to decrease the cost of efficient health management and improve quality of life. The use of a health monitoring observer network not only improves quality of life, but also reduces the cost of health monitoring, making it more sustainable for the general community.

An adapted hourly Himawari-8 fire product for China: principle, methodology and verification

Abstract. Wildfires exert strong influences on the environment, ecology, economy and public security. However, the existing hourly Himawari-8 fire product produced by the Japan Meteorological Agency (JMA) presents large uncertainties and is not suitable for reliable real-time fire monitoring in China. To fill this gap, the National Satellite Meteorological Center (NSMC) proposed an adaptive hourly Himawari-8 fire product for China based on the original Himawari-8 source by employing a dynamical threshold for fire extraction and a database of ground thermal sources. According to the visually extracted reference and consistency check, we found that the NSMC-Himawari-8 fire product effectively removed a majority of false fire alarms included in the original Himawari-8 fire product. Based on a rare field-collected ground reference dataset, we evaluated the reliability of original Himawari-8 and NSMC-Himawari-8 fire products across China. The overall accuracy of the raw Himawari-8 fire product was 54 % and 59 % (not considering the omission errors), respectively. As a comparison, by identifying more real fire pixels and avoiding a majority of false fire alarms, the overall accuracy of the NSMC-Himawari-8 fire product was 80 % and 84 % (not considering the omission errors), respectively, making it an ideal source for improved real-time fire monitoring across China. This research also provides a useful reference for employing a local dataset of underlying surfaces and thermal sources to enhance the accuracy of global fire products in specific regions. The NSMC-Himawari-8 fire product can be downloaded at http://figshare.com (last access: 17 April 2023) with the following DOI: https://doi.org/10.6084/m9.figshare.21550248 (Chen et al., 2022a).

ASSESSMENT OF FIRE RISK IN THE VALE DO PARAÍBA REGION, SOUTHEAST ATLANTIC RAINFOREST, BRAZIL

In recent times, the area burned by wildfires in the Atlantic Rainforest, a biodiversity hotspot, has increased and its occurrence may threaten this biome. The objective of this study was to evaluate the performance of three fire risk indices for a historical time series in Vale do Paraíba Paulista, southeast Atlantic Rainforest, Brazil. Daily meteorological data from automatic weather stations and hotspots records from INPE fire monitoring program were used to validate the formulas. Three fire risk indices were calculated: Angstron, Monte Alegre Formula (MAF) and Telecyn. We expected that we would find an increase in fire risk in recent years in the region, which we found for some municipalities, and that burning risk would be higher in dry months, which we showed. Moreover, we argued that protected areas surrounding higher fire risk sites are being threatened, especially near São Luiz do Paraitinga and Taubaté. Lastly, considering the higher probability in detecting fire risk in higher classes, Angstron was the most adequate for Campos do Jordão and MAF for Sao Luiz do Paraitinga, Taubaté and Cachoeira Paulista.

An Accurate Forest Fire Recognition Method Based on Improved BPNN and IoT

Monitoring and early warning technology for forest fires is crucial. An early warning/monitoring system for forest fires was constructed based on deep learning and the internet of things. Forest fire recognition was improved by combining the size, color, and shape characteristics of the flame, smoke, and area. Complex upper-layer fire-image features were extracted, improving the input conversion by building a forest fire risk prediction model based on an improved dynamic convolutional neural network. The proposed back propagation neural network fire (BPNNFire) algorithm calculated the image processing speed and delay rate, and data were preprocessed to remove noise. The model recognized forest fire images, and the classifier classified them to distinguish images with and without fire. Fire images were classified locally for feature extraction. Forest fire images were stored on a remote server. Existing algorithms were compared, and BPNNFire provided real-time accurate forest fire recognition at a low frame rate with 84.37% accuracy, indicating superior recognition. The maximum relative error between the measured and actual values for real-time online monitoring of forest environment indicators, such as air temperature and humidity, was 5.75%. The packet loss rate of the forest fire monitoring network was 5.99% at Longshan Forest Farm and 2.22% at Longyandong Forest Farm.

Forest Fire Mapping Using Multi-Source Remote Sensing Data: A Case Study in Chongqing

Forest fires are one of the most severe natural disasters facing global ecosystems, as they have a significant impact on ecological security and social development. As remote sensing technology has developed, burned areas can now be quickly extracted to support fire monitoring and post-disaster recovery. This study focused on monitoring forest fires that occurred in Chongqing, China, in August 2022. The burned area was identified using various satellite images, including Sentinel-2, Landsat8, Environmental Mitigation II A (HJ2A), and Gaofen-6 (GF-6). The burned area was extracted using visual interpretation, differenced Normalized Difference Vegetation Index (dNDVI), and differenced Normalized Burnup Ratio (dNBR). The results showed that: (1) The results of the three monitoring methods were very consistent, with a coefficient of determination R2 > 0.96. (2) A threshold method based on the dNBR-extracted burned area was used to analyze fire severity, with moderate-severity fires making up the majority (58.05%) of the fires. (3) Different topographic factors had some influence on the severity of the forest fires. High elevation, steep slopes and the northwestern aspect had the largest percentage of burned area.

Integrating Machine Learning and a Spatial Contextual Algorithm to Detect Wildfire from Himawari-8 Data in Southwest China

Timely wildfire detection is helpful for fire monitoring and fighting. However, the available wildfire products with high temporal resolutions face problems, such as high omission error and commission error (false alarm) rates. This study proposed a wildfire detection algorithm combined with an improved spatial contextual algorithm and machine learning method in southwest China. First, a dataset consisting of a formation of high-confidence fire pixels combining the WLF (Himawari Wild Fire product) and VIIRS wildfire products was constructed. Then, a model to extract potential fire pixels was built using the random forest method. Additionally, an improved spatial contextual algorithm was used to identify actual fire pixels from potential fire pixels. Finally, strategies such as sun glint rejection were used to remove false alarms. As a result, the proposed algorithm performed better, with both a lower omission error rate and a lower commission error rate than the WLF product. It had a higher F1 score (0.47) than WLF (0.43) with VIIRS for reference, which means it is more suitable for wildfire detection.

Основные направления предупреждения чрезвычайных лесопожарных ситуаций

The study investigates the central current issues in extreme forest fire extinction from organizational and technological perspectives. An analysis of the problems in the field of forest pyrology is established on the principles of systematic approach, methods of logical analysis and synthesis of ongoing processes. It indicates the need to develop modern mechanisms against forest fires of various types, which are often the origins of emergency forest fires of various scales. It has been determined that the scientifically substantiated ways of arrangement and implementation of measures for fire prevention and elimination should be supported by an adequate functioning system for scientifically based management decisions. Such a system should be grounded on the processes of the integrated fire monitoring and predicting of the forest fire situation, supervising and estimating the dynamics as well as consequences of the forest fires, their sources of various types and emergency scenarios with gradated threat scales. There are multiple fire extinguishing techniques with various productivity, environmental friendliness, cost efficiency, speed response that have been practically proven over a long period of time. Although the elevation of the results, according to the parameters of the reduced number of burns and the forest areas covered by fire, shows the need for improvement, as well as the generation and realization of the new modern methods against burns on wooded areas especially with the damaging factors. It is recommended to reduce the fire hazard class caused by weather conditions by using an environmentally compatible and efficient electrophysical principals. Such response procedures have gone through an extensive practical examination in different geographical conditions of the Russian Federation and in several foreign countries. It should be mentioned that this approach has a great potential for prevention and termination of large forest fires, reducing the damaging factors coming from the sources of emergency fire situations. For citation: Podrezov Yu.V. Principal Directions for Prevention of Emergency Fire Situations in Forests. Lesnoy Zhurnal = Russian Forestry Journal, 2023, no. 2, pp. 172–182. (In Russ.). https://doi.org/10.37482/0536-1036-2023-2-172-182

Estimation of Root-Zone Soil Moisture in Semi-Arid Areas Based on Remotely Sensed Data

Soil moisture (SM) is a bridge between the atmosphere, vegetation and soil, and its dynamics reflect the energy exchange and transformation between the three. Among SM at different soil profiles, root zone soil moisture (RZSM) plays a significant role in vegetation growth. Therefore, reliable estimation of RZSM at the regional scale is of great importance for drought warning, agricultural yield estimation, forest fire monitoring, etc. Many satellite products provide surface soil moisture (SSM) at the thin top layer of the soil, approximately 2 cm from the surface. However, the acquisition of RZSM at the regional scale is still a tough issue to solve, especially in the semi-arid areas with a lack of in situ observations. Linking the dynamics of SSM and RZSM is promising to solve this issue. The soil moisture analytical relationship (SMAR) model can relate RZSM to SSM based on a simplified soil water balance equation, which is suitable for the simulation of soil moisture mechanisms in semi-arid areas. In this study, the Xiliaohe River Basin is the study area. The SMAR model at the pixels where in situ sites were located is established, and parameters (a, b, sw2, sc1) at these pixels are calibrated by a genetic algorithm (GA). Then the spatial parameters are estimated by the random forest (RF) regression method with the soil, meteorological and vegetation characteristics of the study area as explanatory variables. In addition, the importance of soil, climatic and vegetation characteristics for predicting SMAR parameters is analyzed. Finally, the spatial RZSM in the Xiliaohe River Basin is estimated by the SMAR model at the regional scale with the predicted spatial parameters, and the variation of the regional SMAR model performance is discussed. A comparison of estimated RZSM and in-situ RZSM showed that the SMAR model at the point and regional scales can both meet the RMSE benchmark from NASA of 0.06 cm3·cm−3, indicating that the method this study proposed could effectively estimate RZSM in semi-arid areas based on remotely sensed SSM data.

A weighted contextual active fire detection algorithm based on Himawari-8 data

ABSTRACT Himawari-8, a geostationary satellite, is equipped with the Advanced Himawari Imager (AHI) sensor, which offers significant advantages for forest fire monitoring. This study proposes a weighted contextual fire detection algorithm (AHI_WFDA) that can apply to the AHI sensor as a global fire detection algorithm. Unlike the traditional pass-by screening algorithms, the algorithm takes into account the characteristics of different bands and assigns different weights and corresponding thresholds to the test conditions based on the bands' sensitivity to fire. To validate the algorithm's performance, we tested it on fires in five target areas. We regard MODIS data as the ground truth data and it was used as the benchmark for comparison with the AHI_WFDA , the Himawari-8 official product WLF, and the traditional spatial contextual algorithm (the reproduced SEVIRI algorithm). The results show that the AHI_WFDA significantly reduces the commission error rate compared to the WLF product. While our algorithm's accuracy rate is not superior to the SEVIRI algorithm, it detects more fire incidents correctly. Compared with the MODIS active fire product, the AHI_WFDA's omission error rate is about 63%. In contrast, the relative commission error rate is about 12%, which is in line with the results of some previous studies. In addition, we conducted detailed verification of our algorithm's results with the support of the Landsat series and Sentinel-2 data. The results show that the algorithm in this paper can effectively exploit the fire detection capability of AHI sensors and provide a new idea for the subsequent algorithms.

A Triggerless Backdoor Attack and Defense Mechanism for Intelligent Task Offloading in Multi-UAV Systems

In recent years, multi-unmanned aerial vehicular systems (MUAV) have become prevalent in divergent applications: agriculture, spectrum utilization, transportation, forest fire monitoring, among others, due to their flexible, robust, and autonomous operational maneuver. Battery-powered multi-UAV systems possess limited computation and communication resources, significantly reducing their functional dimension by limiting mission time and range. To address this issue, we propose a federated deep reinforcement learning (FDRL) based intelligent and decentralized task offloading scheme for resource-constrained UAVs that can enhance the operational capability of the MUAV systems. Moreover, the proposed FDRL scheme can improve offloading policy quality while preserving data privacy in MUAV. However, such intelligent systems may fall prey to backdoor attacks that can intervene in the system’s regular operation causing rapid degradation of its performance. We introduce a novel triggerless backdoor attack scheme on intelligent task offloading UAVs and analyze its impact to gauge the resiliency of the offloading policy in the presence of an adversary. Then, we propose lightweight agnostic defense mechanisms to combat such backdoors in multi-UAV settings. The extensive simulation results show that the proposed attack and defense strategies are practical and efficient.

Research on Automatic Patrol Inspection Technology Scheme for Safe Operation of Super Long and Long Span Highway Tunnel

Firstly, this paper conducts systematic research on tunnel fire and traffic incident detection technologies, and develops a high-precision monitoring and detection implementation plan for ultra-long and large-span tunnels; Secondly, in terms of tunnel fire detection, video thermal imaging technology is used as the main detection method, combined with dual wavelength flame detectors, to achieve fire open fire monitoring and pure smoke detection, making up for the blind spots of traditional fire detection, and improving the accuracy, timeliness, and reliability of monitoring. In terms of traffic incident detection, based on traditional traffic incident detection technology and advanced technical means, the system mainly focuses on lightning visual integration detection, combined with artificial video monitoring, to comprehensively judge the status of tunnel traffic incidents, thereby reducing the false alarm rate of incident detection. Finally, based on the technical research scheme, an intelligent automatic patrol robot was developed using the mobile automatic patrol mode. Using this as a carrier, various sensor modules and emergency equipment were built to replace manual 24-hour uninterrupted full coverage patrols, achieving automatic patrol, automatic reporting, analysis, early warning, and emergency response for ultra-long tunnels, reducing the cost of manual patrols and improving the efficiency of patrols, Enhanced emergency response capabilities of operating units.