Detection and Classification of Incoming Ambulance Vehicle using Artificial Intelligence Technology

Traffic congestion increases due to a number of vehicles on the road. Drivers cannot see the arrival of emergency ambulance. Then, personal vehicles block emergency ambulance. As a result, the chance of patient’s dead increases due to such problems. Currently, personal vehicles equipped with Car DVR. This article introduces the implementation of AI for ambulance classification from features of ambulance (Text Ambulance, ambulance lights, Red Cross symbol, Star of life symbol) using Faster-RCNN by TensorFlow. Initially, various image pre-processing techniques are performed. Intersection over Union (IoU) are used for accuracy representation. The results of this research found that the accuracy of classification, star of life symbol give accuracy 98%, red cross symbol 99%, text Ambulance 95.6%, and Ambulance light 97.9%. Our results confirm the proposed system can classify and detection an incoming of Ambulance vehicles.