Javascript must be enabled for the correct page display

Geometric Model Matching on License Plates

Timans, H. (2003) Geometric Model Matching on License Plates. Master's Thesis / Essay, Computing Science.

[img]
Preview
Text
Infor_Ma_2003_HTimans.CV.pdf - Published Version

Download (3MB) | Preview

Abstract

The recognition of license plates from input images often does not give a complete answer. There might be some characters that are not recognized by a neural network, or there might be more characters recognized then possible on a single plate. Using constraints in the syntax of a license plate it is possible to rule out a number of options, but certainly not all of them. To obtain the correct data from these license plates images an evaluation of the plate geometry is required. The goal in this thesis is to chart all the errors that occur in todays license plate recognition software, and use the results to create a license plate model for geometric matching. By using both geometry and syntax properties found in the dataset and the constraints put on license plate geometry and syntax by regulations an attempt is made to achieve a higher rate of recognition then before.

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Computing Science
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 07:29
Last Modified: 15 Feb 2018 07:29
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/8865

Actions (login required)

View Item View Item