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