Driving Alert System through feature extraction on facial images using neural network classifier

Abstract

Any person driving for a long period of time without taking rest causes them to be weary. This driving alert system alerts the driver from causing accidents. It uses Radial Basis Function neural network (RBFNN) for classifying the drivers’ facial expression. This approach extracts the features from facial characteristics points (FCP) of face images. Nearly,19 FCP are selected from the face image at variable lighting condition .The state of the eye is detected by processing frontal or side views of the face image taken by a camera mounted in the car and then automatically trained using a cross correlation based optical flow method. Forty eight features are extracted from the FCP based on openness and closeness of eye, width of mouth. The input features are then categorized into normal, anger and drowsiness expression. The performance is evaluated in terms of recognition rate of RBFNN classifier.

Authors and Affiliations

V. Ezhilarasi, N. Srinivasan, M. Kavitha

Keywords

Related Articles

Effect of Non Linear Impairments on Different Optical Fibers

In this paper the analysis of stimulated raman scattering induced crosstalk has been discussed. The expression for SRS induced crosstalk have been observed at different fiber types such as single mode fiber, dispersion...

Effect of Location on Construction Projects

In this document attempt has been made to study various types of construction projects and the location of these projects in different areas. A comparison has been made by the analysis of effect of location on the const...

Evaluation of Performance and Emission Characteristics of LHR Engine with Biodiesel as Fuel

The challenge for Automobile is present emission norms that demands engine for green environment with high performance and low emission. The Low Heat Rejection Engine (LHR Engine) is a technology, which minimizes heat l...

Use of Renewable Energy Sources in Construction of Green Building

This paper mainly aims towards the green building concept which includes the renewable sources used for the construction. The green building concept is very much useful in keeping the environment clean which on a certai...

Singular-Value-Decomposition Based Generalized Inverse Aided Method PAPR-Aware Linear Precoder Designing for MIMO Based OFDM System

Here we propose a linear precoding scheme for a single user multiple-input–multiple-output orthogonal frequency division multiplexing (OFDM) system to minimize PAPR (peak to average power ratio) by using some redundant...

Download PDF file
  • EP ID EP22955
  • DOI -
  • Views 222
  • Downloads 3

How To Cite

V. Ezhilarasi, N. Srinivasan, M. Kavitha (2016). Driving Alert System through feature extraction on facial images using neural network classifier. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 4(12), -. https://europub.co.uk/articles/-A-22955