Please use this identifier to cite or link to this item:
Title: Vehicle and human tracking adapting different analysis techniques
Authors: Mok, Serene Yin Leng
Keywords: DRNTU::Engineering
Issue Date: 2016
Abstract: Object tracking is one of the main problems countered in the process of perfecting computer vision today. Object tracking is very useful in a variety of applications such as driver assistant program and camera surveillance. Tracking is a process whereby an object is being located and followed in either a real-time or pre-recorded series of image sequences. This has been a very challenging problem as most of the present methods are unable to handle different wide varieties of complex images. Therefore it is very important for a tracking system to be equipped with a versatile adaptive filter, also often known as kernel or a classifier, to be able to differentiate the target’s characteristics from its background for efficient and precise tracking. Currently there are many proposed methods and techniques of object tracking, all having their own strengths and weakness. However it is important to know which proposed solution would be best for tracking human and vehicles, therefore comes the purpose of this paper. Using MATLAB algorithms, this paper will analyze the basis of 3 different object tracking (DSST, Color Visual and KCF) and how it is applied to human and vehicle tracking. It will also explore and evaluate the different types of image analysis techniques and filters used for effective and robust tracking.
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
2.34 MBAdobe PDFView/Open

Page view(s) 50

checked on Sep 30, 2020

Download(s) 50

checked on Sep 30, 2020

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.