Conference paper Open Access

A Multi-Layer Fusion Approach For Real-Time Fire Severity Assessment Based on Multimedia Incidents

Gerasimos Antzoulatos; Panagiotis Giannakeris; Ilias Koulalis; Anastasios Karakostas; Stefanos Vrochidis; Ioannis Kompatsiaris


Citation Style Language JSON Export

{
  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.3701479", 
  "language": "eng", 
  "title": "A Multi-Layer Fusion Approach For Real-Time Fire Severity Assessment Based on Multimedia Incidents", 
  "issued": {
    "date-parts": [
      [
        2020, 
        3, 
        9
      ]
    ]
  }, 
  "abstract": "<p>Shock forest fires have short and long-terms devastating impact on the sustainable management and viability of<br>\nnatural, cultural and residential environments, the local and regional economies and societies. Thus, the utilisation of<br>\nrisk-based decision support systems which encapsulate the technological achievements in Geographical Information<br>\nSystems (GIS) and fire growth simulation models have rapidly increased in the last decades. On the other hand,<br>\nthe rise of image and video capturing technology, the usage mobile and wearable devices, and the availability of<br>\nlarge amounts of multimedia in social media or other online repositories has increased the interest in the image<br>\nunderstanding domain. Recent computer vision techniques endeavour to solve several societal problems with<br>\nsecurity and safety domains to be one of the most serious amongst others. Out of the millions of images that exist<br>\nonline in social media or news articles a great deal of them might include the existence of a crisis or emergency<br>\nevent. In this work, we propose a Multi-Layer Fusion framework, for Real-Time Fire Severity Assessment, based<br>\non knowledge extracted from the analysis of Fire Multimedia Incidents. Our approach consists of two levels: (a)<br>\nan Early Fusion level, in which state-of-the-art image understanding techniques are deployed so as to discover<br>\nfire incidents and objects from images, and (b) the Decision Fusion level which combines multiple fire incident<br>\nreports aiming to assess the severity of the ongoing fire event. We evaluate our image understanding techniques in a<br>\ncollection of public fire image databases, and generate simulated incidents and feed them to our Decision Fusion<br>\nlevel so as to showcase our method&rsquo;s applicability.</p>", 
  "author": [
    {
      "family": "Gerasimos Antzoulatos"
    }, 
    {
      "family": "Panagiotis Giannakeris"
    }, 
    {
      "family": "Ilias Koulalis"
    }, 
    {
      "family": "Anastasios Karakostas"
    }, 
    {
      "family": "Stefanos Vrochidis"
    }, 
    {
      "family": "Ioannis Kompatsiaris"
    }
  ], 
  "type": "paper-conference", 
  "id": "3701479"
}
41
29
views
downloads
All versions This version
Views 4141
Downloads 2929
Data volume 105.9 MB105.9 MB
Unique views 3232
Unique downloads 2626

Share

Cite as