Automatic car identification system

• Main Authors: Khalifa, Othman Omran, Balfaqih, Mohammed, Saad Osman, Bashir, Albagul, Abdulgani
• Format: Conference or Workshop Item
• Language: English
• Published: 2012
• Subjects: TL1 Motor vehicles
• Online Access: http://irep.iium.edu.my/24788/1/2108C.pdf; http://irep.iium.edu.my/24788/; http://www.iium.edu.my/icmae/12/
• Institution: Universiti Islam Antarabangsa Malaysia

Automatic car identification is very useful in traffic monitoring and control access applications that make it become relevant in recent years as a result of increased security awareness. Several computer systems were proposed to identify cars via different features such car plate recognition, car front view features, etc. Vision based car Make and Model Recognition (MMR) can overcome some limitations of the other methods and provides additional security to License Plate Recognition systems. Unfortunately, there is no reliable method based on MMR for side view features. This paper proposes an automatic car identification system via Make and Model Recognition (MMR) based on template matching method. The proposed system consists of three main components which are: car detection, car segmentation, and template matching. The main advantage of this method over conventional car recognition systems -typically based on license plate recognition systems- comes from using make and model car that can’t be easily changed. The system was tested using Matlab where it identified 38 cars out of 40 cars. The accuracy of our system is about 95%.