Vehicle License Plate Localization Techniques

Wong, Felix Siew An (2018) Vehicle License Plate Localization Techniques. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Felix Wong Siew An.pdf Restricted to Registered users only Download (2MB)

Abstract

Nowadays, due to the high number of traffic accidents and high number of toll control system, the need of establishment of defense and prevention on traffic control has encouraged to study a well-known image processing technology, which is License Plate Recognition (LPR). LPR system require to go through four main component, which is acquire the image that contain the license plate, pre-processing the image, character segmentation and recognition. Different countries have different styles could be found in their license plates. Each country has to develop their own LPR system that is suitable for their vehicle license plate format. Due to this reason, most of the LPR system has controlled the environment conditions so that the LPR system able to detect and recognize the license plate character in their country. Therefore, in this project, there would be various type of license plate localization technique proposed to locate the presence of License Plate in an image. Analyze one of the suitable types of License Plate Localization Techniques and carry out the experiment are discussed in this thesis. The system utilizes thresholding, feature extraction and median filter to quickly extract the License Plate region. A set of dataset is used for testing. The proposed system achieved 76.5% accuracy for overall localization of license plate.