Emergency Vehicle Type Classification using Convolutional Neural Network

Abstract

This paper discusses the Convolutional Neural Network (CNN) applied in emergency vehicle image classification. Emergency vehicles are often found stuck in traffic congestion. It has resulted in the emergency vehicles unable to get to the scene quickly. Detecting emergency vehicles on the road can help provide a route to enable emergency vehicles to arrive more efficiently. Several methods have been used to detect the presence of these emergency vehicles on the road. Convolutional Neural Network is one of the popular classification methods nowadays. This work used VGG-16 as the pre-trained model with reduced convolutional layer and filter size. Based on the experiment, the proposed method gained an accuracy of 95%. Thus, the system has achieved the objective.