Automatic license plate detection and recognition / Muhammad Adib Mohd Ariffin
Transportation is important in our daily lives. Nowadays, in Malaysia the usage of vehicle has increased tremendously because of the population growth and human needs. Due to that, Malaysian has produce many types of vehicles to be used. Recently, for each vehicles the license plate has a lot of dif...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://ir.uitm.edu.my/id/eprint/14540/ http://ir.uitm.edu.my/id/eprint/14540/1/TD_MUHAMMAD%20ADIB%20MOHD%20ARIFFIN%20CS%2015_5.pdf |
| Summary: | Transportation is important in our daily lives. Nowadays, in Malaysia the usage of vehicle has increased tremendously because of the population growth and human needs. Due to that, Malaysian has produce many types of vehicles to be used. Recently, for each vehicles the license plate has a lot of different style. Therefore, it is difficult for the authorities to detect and recognize the license plate for security purposes. The objective of this project is to propose a technique that can be used for detection and recognition of license plate. License Plate Recognition (LPR) System is one kind of Intelligent Transport System which can be considered interesting because of its potential application. In the LPR system there are several phases for the detection and recognition of license plate such as image acquisition, pre-processing, segmentation, character segmentation and recognition. For each phase there are technique used to obtain good performance of the license plate detection and character recognition. In this project, connected component analysis for plate recognition and multilayer perceptron neural network (MLPNN) for character recognition is used. From 100 image of vehicles license plate that have been captured for this project, the result for the plate recognition using the required technique is 37% accurate whereas the result for the character recognition using the neural network tool is 100% recognizable from the dataset that is used. The overall result is very satisfying in which most image of license plate captured can be recognize. Finally, this intelligent transportation system is significant in areas of security access control or law enforcement. |
|---|