Date of Publication

11-2019

Document Type

Master's Thesis

Degree Name

Master of Science in Electronics and Communications Engineering

Subject Categories

Electrical and Computer Engineering

College

Gokongwei College of Engineering

Department/Unit

Electronics and Communications Engineering

Thesis Adviser

John Anthony C. Jose

Defense Panel Chair

Edwin Sybingco

Defense Panel Member

Laurence Gan Lim
Elmer P. Dadios

Abstract/Summary

This study can avoid the challenging task of character segmentation and let the neural network learn from the sequence labels, for instance, the vehicle’s plate number. In the case of hybrid graphic-text plates in the Philippines, like the 2003 Rizal plates, preprocessing like binarization, thresholding, component localization will not be required to remove unnecessary objects in the license plates. Since most license plate recognition systems used static images, this study will consider the spatio-temporal information of videos in training and testing the video-object detector or tracker. The tracking module of the proposed neural network model will prevent generating redundant license plate images and recognition results. Improving the current researches in license plate recognition system can help in creating an automatic license plate recognition that can enhance the current implementation of No Contact Apprehension Policy of MMDA. The policy can lessen the traffic congestion caused by flag-down violators since no traffic enforcers will instruct motorists to pull over their car, the corruption and bribery will also be narrowed.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTG007399

Keywords

Image processing—Digital techniques; Intelligent transportation systems

Upload Full Text

wf_yes

Embargo Period

9-16-2022

Download

Share

COinS