Design and development of image processing algorithms for quantitative road traffic data analysis

Most traffic regulators uses surveillance cameras to monitor road traffic patterns, perform post-accident investigation and gather traffic statistics. For statistical data collection, it heavily involves manual counting of vehicles through the surveillance footage which is tedious and unreliable pro...

Full description

Saved in:
Bibliographic Details
Main Author: Ng, Terence Yew Guan.
Other Authors: Mohammed Yakoob Siyal
Format: Final Year Project
Language:English
Published: 2012
Subjects:
Online Access:http://hdl.handle.net/10356/49879
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Most traffic regulators uses surveillance cameras to monitor road traffic patterns, perform post-accident investigation and gather traffic statistics. For statistical data collection, it heavily involves manual counting of vehicles through the surveillance footage which is tedious and unreliable process. With the aid of autonomous traffic analysis software, it improves the process of road traffic data analysis, and it provides more accurate and timely road traffic information. In this project, various image processing techniques were studied and compared. A Matlab GUI was created and programmed to collect statistical data like vehicles classification, vehicles speed, road usage and queue line. Different types of image processing techniques like Background Difference, Edge Detection, Inter-frame difference, Binary image conversion and Quadtree decomposition were used to collect such data. Result from of these techniques will be compared with one another and measured in term of accuracy of the data and timeliness of data collection process.