Abstract

Automatic License Plate Recognition (ALPR) systems capture a vehicles license plate and recognize the license number and other required information from the captured image. ALPR systems have number of significant applications: law enforcement, public safety agencies, toll gate systems, etc. The goal of these systems is to recognize the characters and state on the license plate with high accuracy. ALPR has been implemented using various techniques. Traditional recognition methods use handcrafted features for obtaining features from the image. Unlike conventional methods, deep learning techniques automatically select features and are one of the game changing technologies in the field of computer vision, automatic recognition tasks, natural language processing. Some of the most successful deep learning methods involve Convolutional Neural Networks. This research applies deep learning techniques to the ALPR problem of recognizing the state and license number from the USA license plate. Existing ALPR systems include three stages of processing: license plate localization, character segmentation and character recognition but do little for the state recognition problem. Our research not only extracts the license number, but also processes state information from the license plate. We also propose various techniques for further research in the field of ALPR using deep learning techniques.

Date of publication

Fall 12-1-2015

Document Type

Thesis

Language

english

Persistent identifier

http://hdl.handle.net/10950/303

Share

COinS