[Zurück]

H. Eidenberger:
"Evaluation of content-based image descriptors by statistical methods";
Multimedia Tools and Applications, Volume 5 (2007), Number 3; S. 241 - 258.

Evaluation of visual information retrieval systems is usually performed by
executing test queries and computing recall- and precision-like measures based on
predefined media collections and ground truth information. This process is complex and
time consuming. For the evaluation of feature transformations (transformation of visual
media objects to feature vectors) it would be desirable to have simpler methods available as
well. In this paper we introduce a supplementary evaluation procedure for features that is
founded on statistical data analysis. A second novelty is that we make use of the existing
visual MPEG-7 descriptors to judge the characteristics of feature transformations. The
proposed procedure is divided into four steps: (1) feature extraction, (2) merging with
MPEG-7 data and normalisation, (3) statistical data analysis and (4) visualisation and
interpretation. Three types of statistical methods are used for evaluation: (1) univariate
description (moments, etc.), (2) identification of similarities between feature elements (e.g.
cluster analysis) and (3) identification of dependencies between variables (e.g. factor
analysis). Statistical analysis provides beneficial insights into the structure of features that
can be exploited for feature redesign. Application and advantages of the proposed approach
are shown in a number of toy examples.

Evaluation, Statistical Data Analysis, Feature Design, Visual Information Retrieval, MPEG-7

http://www.springerlink.com/content/r45321015v7706tg/fulltext.pdf