Artificial Intelligence and Applications

Autor: Mohammad Shorif Uddin, Liton Jude Rozario, Mohammad Reduanul Haque
Rok vydání: 2015
Předmět:
Zdroj: Artificial Intelligence and Applications. 2015:1-7
ISSN: 2374-4987
2374-4979
DOI: 10.15764/aia.2015.01001
Popis: Vehicle classification plays an important role in a variety of applications like surveillance, security system, traffic congestion avoidance and accidents prevention etc. However, this is difficult for computer vision to achieve due to variations in shapes and colors of vehicles. In this paper, we present a vision-based vehicle classification method, which has the capability of classifying a wide selection of vehicles. The vehicle classification method consists of two stages ‐ extraction of features using Gabor Filter and then classification using Support Vector Machine (SVM). The method shows good performance in operating on realistic standard data sets and also confirms robust to partial occlusions.
Databáze: OpenAIRE