Disaster Management Applications Research Articles

Cross-Domain Transfer Learning for Natural Scene Classification of Remote-Sensing Imagery

Natural scene classification, which has potential applications in precision agriculture, environmental monitoring, and disaster management, poses significant challenges due to variations in the spatial resolution, spectral resolution, texture, and size of remotely sensed images of natural scenes on Earth. For such challenging problems, deep-learning-based algorithms have demonstrated amazing performances in recent years. Among these methodologies, transfer learning is a useful technique which employs the learned features already extracted from the pre-trained models from large-scale datasets for the problem at hand, resulting in quicker and more accurate models. In this study, we deployed cross-domain transfer learning for the land-cover classification of remotely sensed images of natural scenes. We conducted extensive experiments to measure the performance of the proposed method and explored the factors that affect the performance of the models. Our findings suggest that fine-tuning the ResNet-50 model outperforms various other models in terms of the classification accuracy. The experimental results showed that the deployed cross-domain transfer-learning system achieved outstanding (99.5% and 99.1%) accurate performances compared to previous benchmarks on the NaSC-TG2 dataset with the final tuning of the whole structure and only the last three layers, respectively.

Background-Mixed Augmentation for Weakly Supervised Change Detection

Change detection (CD) is to decouple object changes (i.e., object missing or appearing) from background changes (i.e., environment variations) like light and season variations in two images captured in the same scene over a long time span, presenting critical applications in disaster management, urban development, etc. In particular, the endless patterns of background changes require detectors to have a high generalization against unseen environment variations, making this task significantly challenging. Recent deep learning-based methods develop novel network architectures or optimization strategies with paired-training examples, which do not handle the generalization issue explicitly and require huge manual pixel-level annotation efforts. In this work, for the first attempt in the CD community, we study the generalization issue of CD from the perspective of data augmentation and develop a novel weakly supervised training algorithm that only needs image-level labels. Different from general augmentation techniques for classification, we propose the background-mixed augmentation that is specifically designed for change detection by augmenting examples under the guidance of a set of background changing images and letting deep CD models see diverse environment variations. Moreover, we propose the augmented & real data consistency loss that encourages the generalization increase significantly. Our method as a general framework can enhance a wide range of existing deep learning-based detectors. We conduct extensive experiments in two public datasets and enhance four state-of-the-art methods, demonstrating the advantages of our method. We release the code at https://github.com/tsingqguo/bgmix.

Bipolar intuitionistic fuzzy graph based decision-making model to identify flood vulnerable region.

Bipolar intuitionistic fuzzy graphs (BIFG) are an extension of fuzzy graphs that can effectively capture uncertain or imprecise information in various applications. In graph theory, the covering, matching, and domination problems are benchmark concepts applied to various domains. These concepts may not be defined precisely using a crisp graph when the vertices and edges are more uncertain. Therefore, this study defines the covering, matching and domination concepts in bipolar intuitionistic fuzzy graphs (BIFG) using effective edges with certain important results. To define these concepts when the effective edges are absent, some novel approaches are discussed. To illustrate the domination concepts, the applications in disaster management and location selection problems are discussed. Further, a BIFG-based decision-making model is designed to identify the flood-vulnerable zones in Chennai, where the city's most and least vulnerable zones are identified. From the proposed model, Kodambakkam ([Formula: see text]) is the most susceptible zone in Chennai. Finally, a comparative analysis is done with the existing techniques to show the efficiency of the model.

Ghana’s Disaster Management Role and Application of Information Technology

Ghana is susceptible to disasters including floods, earthquakes, domestic and industrial fires, rain and windstorms, epidemics, pest and insect infestation, boat and road accidents, and many others. There are no clear policies in place to help citizens cope with disasters or even report them if they occur. The research examines historical and current instrumental records, as well as the role of information technology in disaster management in Ghana. Furthermore, the study aims to assist state institutions in understanding the different types of ICT tools on which they can rely at each stage of a disaster to help inform, educate, and mitigate during a disaster. The survey was used to collect primary data from the study population, and the study’s methodology was also qualitative. Floods and wildfires, both domestic and industrial, are common in Ghana, according to the study's findings. It was also discovered that radio is the most widely used ICT tool for disaster reporting.

Disaster and risk management in outdoor recreation and tourism in the context of climate change

PurposeThe increasing frequency and intensity of the extreme weather events could cause devastating consequences in tourism. Climate change–related extreme weather events and their relation to tourism is an emerging field for education and research. The purpose of this study is to categorize the impact of climate change on tourist destinations with regard to extreme weather-related risks in outdoor recreation and tourism. Managerial implications for policymakers and stakeholders are discussed.Design/methodology/approachTo outline the risks from climate change associated with tourism, this study uses the Prisma analysis for identification, screening, checking for eligibility and finding relevant literature for further categorization.FindingsBased on a thoroughly examination of relevant literature, risks and threats posed by climate change could be categorized into following four areas: reduced experiential value in outdoor winter recreation; reduced value in beach scenery and comfort; land degradation and reduced biodiversity; and reduced value in personal safety and comfort in tourism. It also focuses on the significance of using big data applications in catastrophic disaster management and risk reduction. Recommendations with technology and data analytics to continuously improve the disaster management process in tourism education are provided based on findings of this study.Originality/valuePrimary contributions of this study include the following: providing a summarized overview of the risks associated with climate change in terms of tourist experiential value for educational implications; and revealing the role of data analytics in disaster management in the context of tourism and climate change for tourism education.

From science to practical applications in disaster management: views from UN-SPIDER

From science to practical applications in disaster management: views from UN-SPIDER

Making Robotic Swarms Trustful: A Blockchain-Based Perspective

Abstract Lately, the importance of swarm robotics has been recognized in a wide range of areas, including logistics, surveillance, disaster management, agriculture, and other industrial applications. The swarm intelligence introduced by the existing paradigm of artificial intelligence and machine learning often ignores the aspect of providing security and reliability guarantees. Consider a futuristic scenario wherein self-driving cars will transport people, self-driving trucks will carry cargo between warehouses, and a combination of legged robots/drones will ship cargo from warehouses to doorsteps. In the case of such a heterogeneous swarm of robots, it is crucial to ensure a trustful and reliable operating platform for smooth coordination, collaborative decision-making via appropriate consensus, and seamless information sharing while ensuring data security. In this direction, blockchain has been proven to be an effective technology that maintains the transactions (records) in a trustful manner after being validated through consensus. This guarantees accountability, transparency, and trust concerning the storage, safeguarding, and sharing of information among the parties. In this paper, we provide a walkthrough demonstrating the feasibility of using blockchain technology to make the robotic swarm trustful systems in their adoption to critical applications at large-scale. We highlight the pros and cons of the use of cloud vis-a-vis blockchain in swarm robotics. Finally, we present various future research opportunities pertaining to the adoption of blockchain technology in swarm robotics applications.

Enhancing building extraction from remote sensing images through UNet and transfer learning

Performing accurate extraction of buildings from remote sensing (RS) images is a crucial process with widespread applications in urban planning, disaster management, and urban monitoring. However, this task remains challenging due to the diversity and complexity of building shapes, sizes, and textures, as well as variations in lighting and weather conditions. These difficulties motivate our research to propose an improved approach for building extraction using UNet and transfer learning to address these challenges. In this work, we tested seven different backbone architectures within the UNet encoder and found that combining UNet with ResNet101 or ResNet152 yielded the best results. Based on these findings, we combined the superior performance of these base models to create a novel architecture, which achieved significant improvements over previous methods. Specifically, our method achieved a 1.33% increase in Intersection over Union (IoU) compared to the baseline UNet model. Furthermore, it demonstrated a superior performance with a 1.21% increase in IoU over UNet-ResNet101 and a 1.60% increase in IoU over UNet-ResNet152. We evaluated our proposed approach on the INRIA Aerial Image dataset and demonstrated its superiority. Our research addresses a critical need for accurate building extraction from RS images and overcomes the challenges posed by diverse building characteristics.

Remote Sensing Information System in Vietnam and Its Application: A Minireview

This study presents a minireview of remote sensing information systems and their application in Viet Nam. This tool plays a vital role in the management of natural resources, land use planning, environmental monitoring, disaster management, and many other applications. Its widespread application in various fields has helped to improve the efficiency and effectiveness of resource management and environmental protection.

Assessment of Temperature, Precipitation, and Snow Cover at Different Altitudes of the Varzob River Basin in Tajikistan

The analysis of precipitation, snow cover, and temperature based on measured data is important for many applications in hydrology, meteorology, climatology, disaster management, and human activities. In this study, we used long-term historical datasets from the Varzob River Basin (VRB) in Tajikistan to evaluate the trend and magnitudinal changes in temperature, precipitation, and snow cover area in the Anzob (upstream), Maykhura (midstream), and Hushyori (downstream) regions of the VRB using the original Mann–Kendall test, modified Mann–Kendall test, Sen’s slope test, and Pettitt test. The results revealed a decreasing trend in the mean monthly air temperature at Anzob station in the upstream region for all months except January, February, and December between 1960 and 2018 and 1991 to 2018. In each of the three regions, the mean annual temperature indicated a clear upward trend. Seasonal precipitation indicated a large increasing trend in January and February at the Anzob station from 1960 to 2018, but a significant downward trend in April in the upstream, midstream, and downstream regions between 1960 and 1990 and from 1991 to and 2018. In the VRB, almost all stations exhibited a downward trend in annual precipitation across all periods, whereas the upstream region showed a non-significant upward trend between 1960 and 1990. The monthly analysis of snow cover in the VRB based on ground data showed that the maximum increase in snow cover occurs in April at the Anzob station (178 cm) and in March at Maykhura (138 cm) and Hushyori stations (54 cm). The Mann–Kendall test, based on MODIS data, revealed that the monthly snow cover in the VRB increased in April and July while a decrease was recorded in February, September, November, and December from 2001 to 2022. The trend’s stable pattern was observed in March, May, August, and October.

CHALLENGES OF ARTIFICIAL INTELLIGENCE APPLICATION FOR DISASTER RISK MANAGEMENT

Abstract. Artificial Intelligence (AI) has the potential to play a significant role in disaster risk management – from predicting disasters to optimizing response efforts. However, its application in disaster management also presents significant challenges. These challenges include the need for high-quality and diverse data, compatibility with existing systems and technologies, ethical and social implications, and ongoing research and development. In addition, data privacy and security is a critical issue as the use of AI in disaster management often involves the collection and analysis of sensitive information. Addressing these challenges is crucial to ensuring that AI is developed and used in ways that are fair, equitable, and effective in reducing the impact of disasters. The aim of the paper is to provide an analysis of challenges of artificial intelligence application for disaster risk management.

Future projections of storm surges and associated coastal inundation along the east coast of India

Abstract This study aims to analyze several aspects of storm surges and associated coastal inundation along the east coast of India. The current study utilizes historical cyclone tracks over the past five decades and, synthetic tracks are projected for the next 100 years to develop a comprehensive analysis of the storm surges in India. The impact of climate change on cyclone path and intensity is also considered. The ADCIRC model is used to compute storm surge heights and associated coastal inundation for historical and future cyclone tracks. An in-depth analysis is carried out using composite maps explaining the storm surge characteristics at various coastal locations. The novelty of this study lies in the comprehensive analysis of potential storm surges and associated coastal flooding related to exaggerated cyclones that are expected in a changing climate scenario. The outcome of the study is beneficial to storm surge operational centers and disaster management applications.

Robotics Applications in Natural Hazards

Natural hazards not only have great contribution to fatality but also economic loss. Although government has proposed well developed polices to rapidly handle emergencies and systematically organized recovery actions, failures of emergency relief, such as ineffective rescue, can significantly increase the post-hazard death rate. With the advance of artificial intelligence, the use of robots for disaster management applications is a new trend in managing and accessing natural disasters. Disaster response robotics is capable of assisting and replacing rescue teams working in dangerous scenarios, which not only alleviates labor-intensity but also reduces the potential risks associated with rescue personnel. Research on construction automation has advanced, but extensive development is required to reach fully autonomous construction in disaster management and post-disaster recovery. Instead, human robotics collaboration is promising and able to effectively alleviate the knowledge deficit and confusion. In the article, different functions of disaster response robotics are introduced, and technical challenges and future improvements are described.

Cloud-based big data framework towards strengthening disaster risk reduction: systematic mapping

PurposeCloud computing promises dependable services offered through next-generation data centres based on virtualization technologies for computation, network and storage. Big Data Applications have been made viable by cloud computing technologies due to the tremendous expansion of data. Disaster management is one of the areas where big data applications are rapidly being deployed. This study looks at how big data is being used in conjunction with cloud computing to increase disaster risk reduction (DRR). This paper aims to explore and review the existing framework for big data used in disaster management and to provide an insightful view of how cloud-based big data platform toward DRR is applied.Design/methodology/approachA systematic mapping study is conducted to answer four research questions with papers related to Big Data Analytics, cloud computing and disaster management ranging from the year 2013 to 2019. A total of 26 papers were finalised after going through five steps of systematic mapping.FindingsFindings are based on each research question.Research limitations/implicationsA specific study on big data platforms on the application of disaster management, in general is still limited. The lack of study in this field is opened for further research sources.Practical implicationsIn terms of technology, research in DRR leverage on existing big data platform is still lacking. In terms of data, many disaster data are available, but scientists still struggle to learn and listen to the data and take more proactive disaster preparedness.Originality/valueThis study shows that a very famous platform selected by researchers is central processing unit based processing, namely, Apache Hadoop. Apache Spark which uses memory processing requires a big capacity of memory, therefore this is less preferred in the world of research.

Automatic Caption Generation for Aerial Images: A Survey

Aerial images have attracted attention from researcher community since long time. Generating a caption for an aerial image describing its content in comprehensive way is less studied but important task as it has applications in agriculture, defence, disaster management and many more areas. Though different approaches were followed for natural image caption generation, generating a caption for aerial image remains a challenging task due to its special nature. Use of emerging techniques from Artificial Intelligence (AI) and Natural Language Processing (NLP) domains have resulted in generation of accepted quality captions for aerial images. However lot needs to be done to fully utilize potential of aerial image caption generation task. This paper presents detail survey of the various approaches followed by researchers for aerial image caption generation task. The datasets available for experimentation, criteria used for performance evaluation and future directions are also discussed.

Space-Air-Ground Integrated Network for Disaster Management: Systematic Literature Review

The occurrence of any kind of natural disaster will eventually lead to the loss of life and property. Countries where such disasters occur make every effort to monitor such disasters and aid as quickly as possible. However, in some cases, a rescue cannot be sent because no information is available to initiate any type of rescue operation. This is usually because common disaster management systems (DMS) use on board or ground networks to route information from the disaster scene to rescue headquarters (HQ), which in most cases cannot provide the information efficiently. One effective approach is to use satellites in conjunction with existing air-to-ground systems. This study provides a comprehensive and systematic overview of the complexities of the space-air-ground integrated network (SAGIN) in disaster management applications, including different architectures and protocols. The main rationale behind this review is to provide an extensive analysis of existing disaster management systems that are making use of SAGIN. This paper also presents the taxonomy for disaster management systems and challenges. Moreover, this research work also highlights open research issues and challenges for any type of disaster scenario. Our results indicate that several challenges are faced by disaster management systems such as hardware-based challenges, network-based characteristics and communication protocols related challenges, availability and accuracy of imagery data, and security and privacy issues.

Editorial for Special Issue: “Remote Sensing Based Building Extraction II”

Accurate building extraction from remotely sensed images is essential for topographic mapping, urban planning, disaster management, navigation, and many other applications [...]

DisastDrone: A Disaster Aware Consumer Internet of Drone Things System in Ultra-Low Latent 6G Network

Internet of Things application in disaster responses and management is a predominant research domain. The introduction of the consumer drones, flying ad-hoc networks, low latency 5G, and beyond 5G produce significant acceleration of this research. The work proposed the implementation of the Consumer Internet of Drone Things (CIoDT) framework for emergency message transfer among smart grid systems and other power sources and enables stable network connectivity in a disaster scenario. A real-life mobility model has been engineered, and the edge-enabled opportunistic MQTT message transfer mechanism is implemented. Also, a dedicated network slice is devised to examine routing performance. A real-life prototype test-bed emulator has been designed to evaluate mobility, message delivery, and flight attitude performance. The results show a message delivery probability of nearly 0.99 in Quality of Service 2(QoS2), and 1.19 seconds of end-to-end latency in Quality of Service level 1(QoS1). An experiment also shows a 94% of coverage ratio in a network slice for QoS2, 84% of bandwidth utilization per network slice, and a minimum of 100ms latency per network slice. The proposed system is highly suitable for mass production of light weight drone network for disaster management.

Blockchain Driven Metaheuristic Route Planning in Secure Wireless Sensor Networks

Recently, Internet of Things (IoT) has been developed into a field of research and it purposes at linking many sensors enabling devices mostly to data collection and track applications. Wireless sensor network (WSN) is a vital element of IoT paradigm since its inception and has developed into one of the chosen platforms for deploying many smart city application regions such as disaster management, intelligent transportation, home automation, smart buildings, and other such IoT-based application. The routing approaches were extremely-utilized energy efficient approaches with an initial drive that is, for balancing the energy amongst sensor nodes. The clustering and routing procedures assumed that Non-Polynomial (NP) hard problems but bio-simulated approaches are utilized to a recognized time for resolving such problems. With this motivation, this paper presents a new blockchain with Enhanced Hunger Games Search based Route Planning (BCEHGS-RP) scheme for IoT assisted WSN. The presented BCEHGS-RP model majorly employs BC technology for secure communication in the IoT supported WSN environment. In addition, an effective multihop route planning approach was designed by the use of EHGS technique. The proposed EHGS technique is derived from the concept of Hill Climbing strategy (HCS) and HGS algorithm. Moreover, a fitness function with two parameters namely residual energy (RE) and inter-cluster distance to elect optimal routes. The performance validation of the BCEHGS-RP model is experimented with under diverse number of nodes. Extensive experimental outcomes highlighted the better performance of the BCEHGS-RP technique on recent approaches.

Efficient Multimedia Content Transmission Model for Disaster Management using Delay Tolerant Mobile Adhoc Networks

Natural and manmade disasters such as earthquakes, floods, unprecedented rainfall, etc. pose several threats to our society. The citizens upload disaster information in the form of multimedia content such as pictures, audio, and videos. Efficient information and communication framework are critical for disaster management. Mobile Adhoc Networks (MANET) have been used effectively for disaster management. However, Disaster management prerequisites following Quality of Service (QoS) requirements such as bandwidth, high delivery ratio, low overhead, and minimal latency; however, the existing data transmission scheme induces high latency and overhead among intermediate devices; In order to meet the QoS requirement of disaster management applications in this paper, High Delivery Efficiency and Low Latency Multimedia Content Transmission (HDELL-MCT) scheme for MANETs is presented. Then, an improved buffer management scheme is presented for meeting disaster management performance and latency prerequisites. The experiment is conducted using ONE Simulator, the outcome shows the HDELL-MCT scheme achieves very good performance considering different QoS metrics such as improving delivery ratio by 38.02%, reducing latency by 7.53% and minimizing hop communication overhead by 65.1% in comparison with existing multimedia content transmission model.