| | |
Summary: Improvement of Histogram-Based Image Retrieval and Classification
Sven Siggelkow, Hans Burkhardt
Albert-Ludwigs-Universit¨at Freiburg
Institute for Pattern Recognition and Image Processing
79085 Freiburg, Germany
sven.siggelkow hans.burkhardt @informatik.uni-freiburg.de
Abstract
Histogram-based techniques are commonly used in im-
age retrieval ranging from basic color histograms to so-
phisticated histograms of various local feature vectors. Our
approach considers multidimensional histograms of locally
invariant features. In order to get rid of the discontinuous
behavior of traditional histograms we develop a fuzzy his-
togram which does no sharp assignment of a feature vector
to one bin only but performs a weighted assignment to all
neighboring bins. This improvement is not restricted to the
kind of feature histograms considered here but could gen-
erally improve histogram-based retrieval or classification
techniques.
1. Introduction
|
