Site Selection of Traffic Camera based on Dempster-Shafer and Bagging Theory

Traffic incident has bad effect on all parts of society so controlling road networks with enough traffic devices could help to decrease number of accidents, so using the best method for optimum site selection of these devices could help to implement good monitoring system. This paper has considered here important criteria for optimum site selection of traffic camera based on aggregation methods such as Bagging and Dempster-Shafer concepts. In the first step, important criteria such as annual traffic flow, distance from critical places such as parks that need more traffic controlling were identified for selection of important road links for traffic camera installation, Then classification methods such as Artificial neural network and Decision tree algorithms were employed for classification of road links based on their importance for camera installation. Then for improving the result of classifiers aggregation methods such as Bagging and Dempster-Shafer theories were used.

Authors:

• S. Rokhsari
• M. Delavar
• A. Sadeghi-Niaraki
• A. Abed-Elmdoust
• B. Moshiri

Keywords:

• Aggregation
• Bagging theory
• Dempster-Shafer theory
• Site selection

References:

[1] N. Shoukry, "Artificial Neural Network in Classification of Severity
Levels in Crashes with Guardrail", USA, 2005.
[2] L. Dong, "Using Bagging Classifier to Predict Protein Domain
Structural Class", China,2006
[3] L. Breiman, "Bagging Predictors", USA, 2006.
[4] D. Koks, "An Introduction to Bayesian and Dempster-Shafer Data
Fusion", Australia, 2005.
[5] J.Dezert,"Evidential Reasoning for Multi-Criteria Analysis based on
DSmT-AHP", France, 2000.
[6] S.Lehegarat-Mascle,"Application of Dempster-Shafer Evidence Theory
to Unsupervised Classification in Multisource Remote Sensing", IEEE
VOL. 35, NO. 4, 1997.
[7] S.rokhsari,"Optimum site selection of traffic camera for efficient control
of roads to decrease traffic incidents based on AHP and Voronoi
diagram and fusion concepts", ITAC conference, Iran, 2012.