In recent decades, the intersection of natural disasters, urban infrastructure, and human lives has become an increasingly critical area of study. This thesis explores the multifaceted impact of natural disasters on urban environments and the important role of advanced technological interventions, specifically remote sensing, and deep learning, in enhancing disaster response and management. By examining the frequency, severity, and consequences of natural disasters, including the loss of human lives and the economic damages incurred, this work delves into the necessity of rapid and efficient disaster response mechanisms. Natural disasters, from earthquakes and tsunamis to floods and droughts, pose significant threats to urban infrastructure and human safety, causing widespread destruction and displacement. The escalating economic losses associated with these events highlight the urgent need for effective disaster risk reduction strategies and preparedness plans. The thesis underscores the importance of remote sensing technology and deep learning algorithms in improving disaster management practices. Through comprehensive analysis, it demonstrates how these technologies facilitate the accurate assessment of damage, in some cases the prediction of future disasters, and implementation of mitigation and recovery strategies. This work is structured around the exploration of remote sensing fundamentals, including the principles of electromagnetic spectrum utilization and the various platforms and sensors that enable the detailed observation of Earth's surface. It further elaborates on the critical applications of these technologies across the disaster management cycle, encompassing prevention, mitigation, preparedness, response, and recovery phases. By integrating case studies and theoretical perspectives, the thesis presents an understanding of how remote sensing and deep learning contribute to making urban environments more resilient to the impacts of natural disasters. As urban areas continue to expand and the frequency of natural disasters rises, partly due to climate change, the integration of advanced technological solutions in disaster management becomes imperative. This thesis aims to provide an overview of current practices and future directions in leveraging remote sensing and deep learning for disaster response, offering valuable insights for policymakers, engineers, and disaster management professionals. Through this exploration, it seeks to contribute to the ongoing efforts in minimizing the devastating effects of natural disasters on urban infrastructure and human lives, paving the way for safer, more resilient urban environments.
Negli ultimi decenni, la combinazione tra catastrofi naturali, infrastrutture urbane e vite umane è diventata un'area di studio sempre più critica. Questa tesi esplora l'impatto delle catastrofi naturali sull'ambiente urbano e il ruolo importante della tecnologia, nello specifico il telerilevamento e il deep learning, nel migliorare la risposta e la gestione delle catastrofi. Esaminando la frequenza, la gravità e le conseguenze delle catastrofi naturali, inclusa la perdita di vite umane e i danni economici subiti, questo lavoro approfondisce la necessità di meccanismi rapidi ed efficienti di risposta alle catastrofi. Le catastrofi naturali, dai terremoti e maremoti alle inondazioni e siccità, pongono significative minacce alle infrastrutture urbane e alla sicurezza umana, causando distruzioni e trasferimenti siginificativi. Le crescenti perdite economiche associate a questi eventi mettono in luce l'urgente necessità di efficaci strategie di riduzione del rischio e piani di preparazione alle emergenze. La tesi sottolinea l'importanza della tecnologia del telerilevamento e degli algoritmi di deep learning nel migliorare le pratiche di gestione delle catastrofi. Attraverso un'analisi approfondita, si dimostra come queste tecnologie facilitino la valutazione accurata dei danni, in alcuni casi, la previsione di future catastrofi e l'attuazione di strategie di mitigazione e ripristino. Questo lavoro è strutturato attorno all'esplorazione dei fondamenti del telerilevamento, inclusi i principi dell'utilizzo dello spettro elettromagnetico e le varie piattaforme e sensori che consentono l'osservazione dettagliata della superficie terrestre. Si approfondiscono le applicazioni di queste tecnologie rispetto al ciclo di gestione delle catastrofi, che comprende le fasi di prevenzione, mitigazione, preparazione, risposta e ripristino. Integrando casi di studio e teoria, la tesi mostra come il telerilevamento e il deep learning contribuiscano a rendere gli ambienti urbani più resilienti agli impatti delle catastrofi naturali. Poiché le aree urbane continuano a espandersi e la frequenza delle catastrofi naturali aumenta, in parte a causa dei cambiamenti climatici, l'integrazione di soluzioni tecnologiche avanzate nella gestione delle catastrofi diventa imperativa. Questa tesi mira a fornire una panoramica delle pratiche attuali e delle future direzioni nell’applicazione del telerilevamento e del deep learning per la risposta alle catastrofi, offrendo preziosi spunti per i responsabili delle decisioni politiche, gli ingegneri e i professionisti della gestione delle catastrofi. Attraverso questa esplorazione, si mira a contribuire agli sforzi in atto per minimizzare gli effetti devastanti delle catastrofi naturali sulle infrastrutture urbane e sulle vite umane, aprendo la strada ad ambienti urbani più sicuri e resilienti.
Applications of machine learning and remote sensing in disaster management cycle
Hatami Goloujeh, Mehdi
2023/2024
Abstract
In recent decades, the intersection of natural disasters, urban infrastructure, and human lives has become an increasingly critical area of study. This thesis explores the multifaceted impact of natural disasters on urban environments and the important role of advanced technological interventions, specifically remote sensing, and deep learning, in enhancing disaster response and management. By examining the frequency, severity, and consequences of natural disasters, including the loss of human lives and the economic damages incurred, this work delves into the necessity of rapid and efficient disaster response mechanisms. Natural disasters, from earthquakes and tsunamis to floods and droughts, pose significant threats to urban infrastructure and human safety, causing widespread destruction and displacement. The escalating economic losses associated with these events highlight the urgent need for effective disaster risk reduction strategies and preparedness plans. The thesis underscores the importance of remote sensing technology and deep learning algorithms in improving disaster management practices. Through comprehensive analysis, it demonstrates how these technologies facilitate the accurate assessment of damage, in some cases the prediction of future disasters, and implementation of mitigation and recovery strategies. This work is structured around the exploration of remote sensing fundamentals, including the principles of electromagnetic spectrum utilization and the various platforms and sensors that enable the detailed observation of Earth's surface. It further elaborates on the critical applications of these technologies across the disaster management cycle, encompassing prevention, mitigation, preparedness, response, and recovery phases. By integrating case studies and theoretical perspectives, the thesis presents an understanding of how remote sensing and deep learning contribute to making urban environments more resilient to the impacts of natural disasters. As urban areas continue to expand and the frequency of natural disasters rises, partly due to climate change, the integration of advanced technological solutions in disaster management becomes imperative. This thesis aims to provide an overview of current practices and future directions in leveraging remote sensing and deep learning for disaster response, offering valuable insights for policymakers, engineers, and disaster management professionals. Through this exploration, it seeks to contribute to the ongoing efforts in minimizing the devastating effects of natural disasters on urban infrastructure and human lives, paving the way for safer, more resilient urban environments.File | Dimensione | Formato | |
---|---|---|---|
submission_last_version_19.pdf
accessibile in internet per tutti
Dimensione
39.4 MB
Formato
Adobe PDF
|
39.4 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/219415