by Gottfried, Björn, Schuldt, Arne and Herzog, Otthein
Abstract:
In content-based image retrieval we are faced with continuously growing image databases that require efficient and effective search strategies. In this context, shapes play a particularly important role, especially as soon as not only the overall appearance of images is of interest, but if actually their content is to be analysed, or even to be recognised. In this paper we argue in favour of numeric features which characterise shapes by single numeric values. Therewith, they allow compact representations and efficient comparison algorithms. That is, pairs of shapes can be compared with constant time complexity. We introduce three numeric features which are based on a qualitative relational system. The evaluation with an established benchmark data set shows that the new features keep up with other features pertaining to the same complexity class. Furthermore, the new features are well-suited in order to supplement existent methods.
Reference:
Gottfried, Björn, Schuldt, Arne and Herzog, Otthein, "Extent, Extremum, and Curvature: Qualitative Numeric Features for Efficient Shape Retrieval", In KI2007, Springer-Verlag, no. 4667, Osnabrück, Germany, pp. 308–322, 2007.
Bibtex Entry:
@INPROCEEDINGS{Gottfried2007c,
author = {Gottfried, Bj{\"o}rn and Schuldt, Arne and Herzog, Otthein},
title = {{Extent, Extremum, and Curvature}: {Qualitative Numeric Features
for Efficient Shape Retrieval}},
booktitle = {KI2007},
year = {2007},
editor = {Hertzberg, Joachim and Beetz, Michael and Englert, Roman},
number = {4667},
series = {Lecture Notes in Artificial Intelligence},
pages = {308--322},
address = {Osnabr{\"u}ck, Germany},
month = {September10--13},
publisher = {Springer-Verlag},
abstract = {In content-based image retrieval we are faced with continuously growing
image databases that require efficient and effective search strategies.
In this context, shapes play a particularly important role, especially
as soon as not only the overall appearance of images is of interest,
but if actually their content is to be analysed, or even to be recognised.
In this paper we argue in favour of numeric features which characterise
shapes by single numeric values. Therewith, they allow compact representations
and efficient comparison algorithms. That is, pairs of shapes can
be compared with constant time complexity. We introduce three numeric
features which are based on a qualitative relational system. The
evaluation with an established benchmark data set shows that the
new features keep up with other features pertaining to the same complexity
class. Furthermore, the new features are well-suited in order to
supplement existent methods.},
doi = {10.1007/978-3-540-74565-5{\_}24},
isbn = {978-3-540-74564-8},
owner = {pmania},
timestamp = {2012.11.06},
url = {http://www.tzi.de/~aschuldt/publications-ki2007.html}
}