Automatic vehicle classification program using image processing for electronic tollway systems

• Main Authors: Nunez, Kristina D., Santiago, Jose Carlo C., Villamayor, Renato Q., Yu Ping Kun, Desiree D.
• Format: text
• Language: English
• Published: Animo Repository 2001
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/11695
• Institution: De La Salle University

This thesis is aimed at developing an Automatic Vehicle Classification (AVC) System designed to aid and improve on the existing Computerized Toll Collection (CTC) System. Human errors are avoided by automating the vehicle classification process, and thus helps the toll companies minimize, if not eliminate, deficits. AVC Systems depend mainly on image processing. The system takes a digital picture of the vehicle on the lane, processes this information, and determines whether the vehicle is categorized as Class 1 (cars, jeeps, pick-up and vans), Class 2 (buses, small trucks and Class 1 vehicles with 1-axle or 2-axle tracks) or Class 3 (large to long haul trucks). Along with the development of the AVC System is the design of a CTC System that is similar to that developed by Micrologic Inc. The AVC System is integrated into the CTC System to replace the manual input of the vehicle class that is the original task of the toolbooth operator. A further enhancement such as network security in the form of file encryption is also developed. Preliminary testing was conducted using images of actual and small-scale vehicles captured outdoors. Data was gathered using these captured images to determine the criteria for detecting each vehicle class. Actual demonstration, however, is performed using small-scale vehicles as subjects. Since the image-processing program is highly sensitive to ambient light variations and the actual color of the vehicle, errors are likely to occur in the detection process. However, these errors are kept within acceptable limits through the introduction of external factors such as proper lighting and roofing, thus ensuring the reliability of the system at all times of the day.