Using a convolutional neural network to detect emergency vehicles using video cameras in heavy traffic

Zikratov Igor Alexeyevich, Shakhsuvarova Ilona Vitalievna

An automated system for detecting emergency vehicles from surveillance camera recordings using a deep convolutional neural network was proposed. The developed method has shown good results in detecting and classifying ambulances, repair and fire trucks. This technology can be used in an automated system that can detect a special vehicle on a high-traffic road and notify the dispatcher or automatically redirect other vehicles to make way.

Keywords:emergency vehicles, convolutional neural network, automated system, high-traffic road, detecting