Plate recognition for Malaysian vehicles using stroke analysis technique

Abstract

The Road Transport Department of Malaysia has endorsed a specification for vehicle plates that includes the font and size of characters that must be followed by car owners. However, there are a special plate number where this specification is not followed such as Proton, BAMbee, Putrajaya, Tiara, Satria and Perodua. This will cause problems in the recognition phase because existing systems will find difficulty in recognizing these plates. Therefore, this project is aimed of an implementing a recognition system that is capable of solving the mentioned issues using stroke analysis technique. The system is an offline system where the vehicle image is loaded manually from a directory. The loaded image is then pre-processed using image processing techniques. Consequently, the image is converted into a binary image. The plate region is extracted prior to characters extraction. All the characters then undergo thinning process before stroke analysis performs tracing and recognition of the characters. The system displays the output in readable text. The performance analysis has shown that the system is able to recognize Malaysian vehicle plate with more than 95 percent accuracy.