Please use this identifier to cite or link to this item:
Title: Human detection and tracking in surveillance videos
Authors: Ren, Yi.
Keywords: DRNTU::Engineering
Issue Date: 2013
Abstract: This project aims to evaluate current human and object detection and tracking methods in surveillance video systems. Since foreground detection is the foremost requirement for tracking and other further studies, the attention of the project is on foreground detection. From the previous researches, four well-known detection methods have been analyzed, namely frame difference, OpenCV Gaussian mixture model, adaptive Gaussian mixture model and optical flow. At least one algorithm developed from each method has been implemented with the same dataset, in order to compare the performance of the methods. The dataset of this project consists of indoor dataset and outdoor dataset to provide a comprehensive analysis. From the result, BackgroundSubtractorMOG, one of the Gaussian mixture model methods, is suggested to be applied in a real-life surveillance video system. Because it is robust to background changes and it has a relatively fast processing speed.
Schools: School of Electrical and Electronic Engineering 
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
3.58 MBAdobe PDFView/Open

Page view(s)

Updated on Jun 21, 2024


Updated on Jun 21, 2024

Google ScholarTM


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