Date of Award

Spring 2013

Project Type

Thesis

Program or Major

Information Technology

Degree Name

Master of Science

First Advisor

Michael Jouas

Abstract

A license plate, referred to simply as a plate or vehicle registration plate, is a small plastic or metal plate attached to a motor vehicle for official identification purposes. Most governments require a registration plate to be attached to both the front and rear of a vehicle, although certain jurisdictions or vehicle types, such as motorcycles, require only one plate, which is usually attached to the rear of the vehicle.

We present analysis of Automatic License Plate Recognition (ALPR) of New Hampshire (NH) plates using open source products. This thesis contains an implementation of a demonstrated model and analysis of the results. In this paper, OpenCV (computer vision library) and Tesseract (open source optical character reader) is presented as a core intelligent infrastructure. The thesis explains the mathematical principles and algorithms used for number plate detection, processes of proper characters segmentation, normalization and recognition. A description of the challenges involved in detecting and reading license plate in NH, previous studies done by others and the strategies adopted to solve them is also given.

Download

Share

COinS