Vehicle locator system for multi-storage car park management

Automatic recognition of car license plate number became a very important in our daily life because of the unlimited increase of cars and transportation systems, thus it is impossible to manage and monitored by humans manually. Unable to locate parked car always become main problem especially in huge car parking system. Many researchers suggest and employ plate recognition system (which is usually called as Automatic Number Plate Recognition (ANPR)) in order to recognise car. There are a number of different techniques which have been used for recognition of registration plate characters due to the diversity of plate formats, different scales, rotations and non-uniform illumination conditions during image acquisition. Malaysia also has different types of plate number. The development of the ANPR system is still lacking and most of the time they just plug and play of ANPR engine without further evaluation and analysis. Hence, the main aim of this project is to develop ANPR system for common Malaysian plate styles with comprehensive evaluations. Four main objectives of this study which is i) to collect, gather and acquire images of cars in parking, ii) to design ANPR system for Malaysian plate styles, iii) to evaluate the performance of the developed ANPR system, and iv) to develop Graphical User Interface (GUI) for ANPR system to ease of use. The system is developed using three stages of processing: i) plate extraction - identify plate regions, ii) character segmentation - one-by-one character detection and isolation, and iii) character recognition - character matching process. The experimental results show that the system is able to recognise plate number by at least more than 80% of recognition accuracy and less than 1% segmentation errors. The system is ready to be used to detect the presence of cars in parking space