Blob Modification In Counting Vehicles Using Gaussian Mixture Models Under Heavy Traffic

Blob Modification In Counting Vehicles Using Gaussian Mixture Models Under Heavy Traffic

Vol. 10 No.162015/ARPN Journal of Engineering and Applied Sciences (scopus Index)

(Indrabayu, Basri, A. Achmad)

Abstract

This study uses Blob Analysis technique to optimize Gaussian Mixture Model method performance in detecting and counting cars and motorcycles under heavy traffic conditions. It is profound that by optimizing the minimum and maximum blob area in order to obtain proper blob size from a video image’s region of interest will also improve the accuracy of system. The result shows that appropriate parameter settings can increase accuracy by 28.02% for motorcycles and 10.84% for cars.

Keywords: intelligent transportation system, gaussian mixture models, blob analysis, heavy traffic condition.