Deep Learning for Emergency Vehicle Identification: A YOLOv8-Based Approach for Smart City Solutions

Abstract

Accurately identifying emergency vehicles is vital in intelligent transportation systems to ensure prompt response and prioritization. This article introduces a new method that utilizes deep learning algorithm like  YOLOv8 (You Only Look Once, version 8) to detect emergency vehicles in busy traffic situations in real-time. YOLOv8's upgraded design with better feature extraction and faster inference speeds offers a strong solution for detecting emergency vehicles in different lighting, occlusion, and weather conditions. The model has been trained on a varied dataset containing different kinds of emergency vehicles like ambulances, fire trucks, and police cars. The results from the experiment show that YOLOv8 outperforms previous YOLO versions and other top object detection models in terms of precision, recall, and real-time inference. This study emphasizes the capability of YOLOv8 for creating smart city solutions, where quick identification of emergency vehicles can decrease response times and enhance road safety as a whole. The method being suggested also deals with problems concerning incorrect identifications and overlooked detections, guaranteeing increased dependability in various urban environments. This study emphasizes the capability of YOLOv8 in creating smart city solutions, where quick identification of emergency vehicles can reduce response times and enhance road safety.