Image processing based vehicle's license plate tracking and recognition system

Lorita Angeline (2013) Image processing based vehicle's license plate tracking and recognition system. Masters thesis, Universiti Malaysia Sabah.

	Text 24 PAGES.pdf Download (446kB)
	Text FULLTEXT.pdf Restricted to Registered users only Download (3MB)

Abstract

The main objective of this research is to design and develop an automatic license plate localisation and recognition system for vehicle in real street scene with constraints on extrinsic imaging parameters such as pose, illumination, scale, position and varying angles. As the number of applications such as the surveillance system for license plate recognition system increase, there is a need for an accurate system that can be applied to dynamic vehicles in the real street scene. However, one major difficulty in object detection is the fact that different views of the same object can look very different, as an image is essentially the projection of a 3D object onto a 2D plane. The presence of an extraneous moving vehicle’s license plate in the system’s field of view also may affect the processing results. Therefore, a new algorithm for automatic license plate localisation and recognition (ALPR) is proposed on the basis of Signature Analysis and Isotropic Dilation (SAICCA). The localised license plate region is further processed and the characters are segmented into individual character. Properties obtained from the Signature Analysis such as ratio of the character and features extracted from the segmented character is fed and trained in a multi-layer feed-forward back-propagation neural network. Designing and developing of the ALPR system is programmed in MATLAB m-file. Evaluations and assessments have been carried out and observational results show an improvement of 32% in license plate localisation by using Signature Analysis and the improved Connected Component Analysis (SAICCA) compared to the conventional Connected Component Analysis (CCA).

Item Type:	Thesis (Masters)
Keyword:	Automatic License Plate Recognition, Signature Analysis, Isotropic Dilation, SAICCA algorithm, License plate localisation, Computer vision, Neural network
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1-2040 Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Department:	SCHOOL > School of Engineering and Information Technology
Depositing User:	DG MASNIAH AHMAD -
Date Deposited:	16 May 2025 13:12
Last Modified:	16 May 2025 13:12
URI:	https://eprints.ums.edu.my/id/eprint/43793

Actions (login required)

View Item