Development of obstruction detection system using computer vision for generation of data for video analytics
Date of Publication
9-2018
Document Type
Master's Thesis
Degree Name
Master of Science in Electronics and Communications Engineering
Subject Categories
Computer Engineering
College
Gokongwei College of Engineering
Department/Unit
Electronics and Communications Engineering
Thesis Adviser
Edwin Sybingco
Defense Panel Chair
Elmer P. Dadios
Defense Panel Member
Argel A. Bandala
John Antholny C. Jose
Abstract/Summary
This study investigates the use of a multiple cameras for intelligent transportation systems, in which the researcher developed an obstruction detection system for traffic light intersections using computer vision and image processing. The study covers the gathering of data that was used to develop the algorithm. The system is composed of the following components: obstruction detection, license plate detection and optical character recognition. Two video inputs were simultaneously processed to perform the intended functions of the system. The study was implemented through Python programming and its libraries. When an obstruction is detected on the region of interest during the stop cycle of the traffic light, the two cameras captures the current frame and then saves it to a directory. The images where the license plates are visible are then processed to determine the license plate location and and prepared for optical character recognition. The system summarizes the data gathered by creating a csv file with the path to the directory of the frames captured as well as the recognized number plate characters.
Abstract Format
html
Language
English
Format
Electronic
Accession Number
CDTG008109
Keywords
Computer vision; Video surveillance
Upload Full Text
wf_no
Recommended Citation
Mendoza, D. M. (2018). Development of obstruction detection system using computer vision for generation of data for video analytics. Retrieved from https://animorepository.dlsu.edu.ph/etd_masteral/7035
Embargo Period
1-7-2025
