Authors: Žunić, Joviša 
Affiliations: Mathematical Institute of the Serbian Academy of Sciences and Arts 
Title: Shape Descriptors for Image Analysis
Series/Report no.: Zbornik Radova
First page: 5
Last page: 38
Related Publication(s): SELECTED TOPICS ON IMAGE PROCESSING AND CRYPTOLOGY
Issue Date: 2012
ISBN: 978-86-80593-47-0
ISSN: 0351-9406
URL: http://elib.mi.sanu.ac.rs/files/journals/zr/23/zunic2.pdf
Abstract: 
We give an overview of the shape based techniques used in object matching, object identification and object classification tasks. We distinguish between the area based methods, which use all the shape points, and boundary based methods, which use boundary information only. We also discuss a recent ‘multi-component shape’ approach. This approach considers a group of objects as a single but compound object. The idea is already shown to be very efficient in a wide spectrum of applications.
Illustrative examples are provided, including those related to personal signature identification and outliers detections, which have, pretty much, obvious and straightforward applications in security and crime prevention related tasks.
Keywords: image processing | computer vision | shape analysis | object identification | data security
Publisher: Mathematical Institute of the Serbian Academy of Sciences and Arts
Project: Advanced Techniques of Cryptology, Image Processing and Computational Topology for Information Security 
Development of new information and communication technologies, based on advanced mathematical methods, with applications in medicine, telecommunications, power systems, protection of national heritage and education 

Show full item record

Google ScholarTM

Check

Altmetric


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