iRepository at Perpustakaan UniMAP >
The Library >
Conference Papers >
Please use this identifier to cite or link to this item:
|Title: ||Moving vehicle identification using artificial neural network|
|Authors: ||Paulraj, Murugesa Pandiyan, Assoc. Prof. Dr.|
Abdul Hamid, Adom, Assoc. Prof. Dr.
Siti Marhainis, Othman
Sundararaj, Sathish Kumar
|Keywords: ||Hearing impaired;Backpropagation;Multilayer Perceptron (MLP) Neural Network|
|Issue Date: ||27-Feb-2012|
|Publisher: ||Universiti Malaysia Perlis (UniMAP)|
|???metadata.dc.publisher.department???: ||School of Mechatronic Engineering|
|Series/Report no.: ||Proceedings of the International Conference on Man-Machine Systems (ICoMMS 2012)|
|Abstract: ||Hearing impaired people cannot distinguish the sound
from a moving vehicle approaching from their behind. They often
face a risky situation while they are in outdoors. In this paper, a
simple algorithm is proposed to classify the type and distance of
the moving vehicles based on the sound signature recorded from
the vehicles. A simple experimental protocol is designed to record
the vehicle sound under different environment conditions and
velocity of vehicles. The noises emanated from moving vehicles
along the roadside were recorded along with the type and distance
of the vehicle. Autoregressive modeling algorithm is used for the
analysis to extract features from the recorded sound signal. Two
simple Multilayer Perceptron (MLP) models are developed and
trained using Backpropagation algorithm to classify the vehicle
type and its distance. The effectiveness of the network is validated
|Description: ||International Conference on Man Machine Systems (ICoMMS 2012) organized by School of Mechatronic Engineering, co-organized by The Institute of Engineer, Malaysia (IEM) and Society of Engineering Education Malaysia, 27th - 28th February 2012 at Bayview Beach Resort, Penang, Malaysia.|
|Appears in Collections:||Siti Marhainis Othman|
Paulraj Murugesa Pandiyan, Prof. Dr.
Abdul Hamid Adom, Prof. Dr.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.