Please use this identifier to cite or link to this item:
Title: Comparison of Various Techniques for Emergency Vehicle Detection using Audio Processing
Authors: Jonnadula, Eshwar Prithvi
Khilar, Pabitra Mohan
Keywords: VANETs
Emergency Vehicles
Audio Sensing
Machine Learning
Issue Date: Feb-2019
Citation: 1st International Conference on Machine Learning, Image Processing, Network Security and Data Sciences (MIND-2019), Kurukshetra, India, 3 - 4 March, 2019.
Abstract: VANETs is an important part of wireless networking. Vehicular movement is unending expanding wherever on the planet and can cause horrible activity clog. The greater part of the signals till date include a settled green signal arrangement, so the green signal timing is done without considering emergency vehicles. In this way, emergency vehicles, for example, are stuck in congested driving conditions and postponed in achieving their goal can prompt loss of property and important lives. In this paper, we do a comparative study of different methods which are used in identifying the emergency vehicle present on the road. This identification of emergency vehicles is useful in the development of smart cities. This is tried on genuine dataset got from Google Audio Dataset which had obtained and recorded in urban avenues which include distinctive activities and noises like people talking, vehicles horns etc. We found out that an artificial neural network consisting of three hidden layers give the highest accuracy which is 3% more than one hidden layer ANN.
Description: Copyright of this document belongs to proceedings publisher.
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2019_MIND_PMKhilar_Comparision.pdfPaper457.94 kBAdobe PDFView/Open    Request a copy

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.