Automatic License Plate Recognition and Search Engine with AI Technology: A Case Study of Chalong Rat Expressway

Abstract

The Expressway Authority of Thailand (EXAT) has developed the expressway toll collection system into a multilane free-flow (M Flow) toll collection system to enable expressway users to pass through the toll gates quickly without interruption, reducing congestion in front of the toll gates. One of the important factors for the development is identifying the vehicles by using artificial intelligence (AI) technology in the form of Automatic License Plate Recognition (ALPR) for vehicle inspection and vehicle/user identification. And if it is necessary to follow up, make a claim, or collect the toll fees retroactively, the system can automatically
issue a transaction report. This will help reduce the workload of EXAT staff. In addition, in the case of external agencies requesting related information to support their operations. EXAT has then developed the ALPR system on the Chalong Rat Expressway as a pilot route. This research compares accuracy between ALPR system and Ground truth. Accuracy rate of ALPR system (97.5%), Color identification (96%) and Vehicle brand (93%). The developed system also initiates a program for detecting and searching for license plates for the Chalong Rat Expressway through the form of a Web Application which is convenient to use without having to
install a program.