Malaysia's Car Plate Recognition System

Ng, Chun Chet (2017) Malaysia's Car Plate Recognition System. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Ng Chun Chet_FULL TEXT.pdf Restricted to Registered users only Download (2MB)

Abstract

Malaysia’s car plate recognition system is developed to recognize the cars’ license plates with invariant to different varieties such as font type, font size, organization of the characters; and as well as dealing with the unintended objects. The main objective of the proposed system is to automate the recognition and interpretation of the car’s license plate by applying computer vision technology. The proposed system can be extended to serve various purposes such as tracking of cars’ arriving and leaving activities, record the duration of staying, parking fees calculation, security measurements, and many others. The robustness of the proposed system is ensured by the advancement of the four main processes in the entire system workflow which are image pre-processing, license plate localization, character segmentation and optical character recognition. Image pre-processing focused on filtering the unintended object and noises from the image before the localization of the license plate is executed. The process continues with identifying and extract the segment of each character in the license plate and lastly, optical character recognition is applied to convert the character image into machine encoded text. Our proposed method is tested with real-life data and achieved significant results compared to the existing works.