Přístupnostní navigace
E-application
Search Search Close
Publication detail
ŠKORPIL, V. ŠŤASTNÝ, J.
Original Title
Comparison Methods for Object Recognition
Type
conference paper
Language
English
Original Abstract
This paper describes our new unpublished results and outputs of selected methods for object recognition. It is continued to our former research. We are focused to the identification with the aid of moments, with the aid of syntactical analysis and with the aid of neural network algorithms. Momentum method is very sensitive to entry image quality. Syntactic analysis is suitable for rotated objects, high-speed classification and for small changes in the segment edge. Neural network algorithms can be used for high-speed classification with randomly rotated objects and for some differences between learned etalons and classified objects.
Keywords
Neural network, Invariant moments, Primitives, Syntactical analysis, Back-propagation algorithm.
Authors
ŠKORPIL, V.; ŠŤASTNÝ, J.
RIV year
2009
Released
21. 7. 2009
Publisher
WSEAS
Location
Rhodos, Greece
ISBN
978-960-474-097-0
Book
Proceedings of the 13th WSEAS International Conference on Systems
Edition
Edition number
1
Pages from
607
Pages to
610
Pages count
4
BibTex
@inproceedings{BUT29438, author="Vladislav {Škorpil} and Jiří {Šťastný}", title="Comparison Methods for Object Recognition", booktitle="Proceedings of the 13th WSEAS International Conference on Systems", year="2009", series="WSEAS", number="1", pages="607--610", publisher="WSEAS", address="Rhodos, Greece", isbn="978-960-474-097-0" }