Extent, Extremum, and Curvature: Qualitative Numeric Features for Efficient Shape Retrieval (bibtex)
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}
}
Powered by bibtexbrowser