By Topic

Road Visual Condition Rating Assessment Using Artificial Neural Network and Support Vector Machine

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Fajardo, E.L.J. ; Univ. of the Philippines, Diliman, Philippines

Annually, the Philippine government'sDepartment of Public Works and Highways (DPWH) gathers visual road condition data for use in the rehabilitation of the Philippine national road network. Alongside the condition data of the road, inventory and traffic data are also gathered simultaneously. In literature, Visual Condition Index (VCI) is used to determine the condition of a road. The data gathering that will result for the VCI is a lengthy and tedious process, subject to various validations by DPWH, before being considered for budget preparation analysis. This study applies Support Vector Machines (SVM) and Artificial Neural Network (ANN) as methods for determining the visual condition of roads on an inventory and traffic data set with 16continuous-valued and 2 nominal-valued dimensions. These methods were applied on predetermined subsets of the dimensions using 10-fold cross-validation on each subset. The results show that ANN generally performs better classification compared to SVM, the difference within the 95% confidence interval. The best known results reported by ANN accurately predict the road condition by 42%. The results show that the inclusion of a large number of dimensions not directly involve din the computation of the VCI in building the classifier contributed to the low accuracy rating.

Published in:

Modelling Symposium (AMS), 2011 Fifth Asia

Date of Conference:

24-26 May 2011