Please use this identifier to cite or link to this item: http://ktisis.cut.ac.cy/handle/10488/7003
Title: Advanced statistical and adaptive threshold techniques for moving object detection and segmentation
Authors: Marques, Oge
Kasparis, Takis 
Christodoulou, Lakis
Keywords: Computer vision
Statistics
Standard deviations
Computer graphics
Issue Date: 2011
Publisher: IEEE
Source: 17th International Conference on Digital Signal Processing, 2011, Pages 1-6
Abstract: The current research project proposes advanced statistical and adaptive threshold techniques for video object detection and segmentation. We present new statistical adaptive threshold techniques to show the advantages, and how these algorithms overcome the limitations and the technical challenges for object motion detection. The algorithm utilizes statistical quantities such as mean, standard deviation, and variance to define a new adaptive and automatic threshold based on two-frame and three-frame differencing. The proposed algorithms were compared with classic statistical thresholding methods on a testing video for human motion detection, and the experimental results show the effectiveness of the algorithms. Furthermore this research shows an evaluation and comparison among all statistical and adaptive algorithms and proves the benefits of the proposed algorithm.
URI: http://ktisis.cut.ac.cy/handle/10488/7003
ISBN: 978-145770274-7
DOI: 10.1109/ICDSP.2011.6004875
Rights: © 2011 IEEE
Appears in Collections:Κεφάλαια βιβλίων/Book chapters

Show full item record

Page view(s)

18
Last Week
0
Last month
3
checked on Aug 21, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.