In the last decade, over two billion people have become victims of mass casualty incidents (MCIs). The success of the emergency response to such events depends heavily on efficient emergency vehicle transportation. During a MCI, the status of transportation pathways is constantly fluctuating, making it difficult to evaluate real-time traffic delays. Standard traffic delay software is not reliable in routing emergency resources to incident sites. Thus, the question of how can image recognition and deep-learning be used in real-time to aid emergency vehicles in MCI efforts emerges. The “You-Only-Look-Once” method is applied to provide accurate vehicle detection; a convex hull is implemented to conduct road detection; and simple supporting methods are used to describe traffic states. This combination yields a classification of traffic congestion based on defined parameter thresholds. The resulting output will ultimately guide a decision-maker or supplemental model to optimize emergency vehicle deployment.
Object Detection Using Artificial Intelligence: Predicting Traffic Congestion to Improve Emergency Response to Mass Casualty Incidents
Lect. Notes in Networks, Syst.
International Conference on Applied Human Factors and Ergonomics ; 2021 July 25, 2021 - July 29, 2021
Advances in Artificial Intelligence, Software and Systems Engineering ; Kapitel : 36 ; 287-294
08.07.2021
8 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Satellite Communications for Management of Mass Casualty Incidents: The e-Triage Project
British Library Conference Proceedings | 2010
|