Image retrieval using steerable wavelets

ALBERT-LUDWIGS-UNIVERSITÄT FREIBURG
INSTITUTE FOR COMPUTER SCIENCE
Chair of Pattern Recognition and Image Processing
Prof. Dr.-Ing. Hans Burkhardt

Georges-Köhler-Allee 52, Room 01-029,
D-79110 Freiburg, Tel. 0761-203-8260

The Chair of Pattern Recognition and Image Processing is currently offering a

Diplom-/Studienarbeit
(Master thesis/Student research project)

Image retrieval using steerable wavelets

Wavelet analysis represents an image (signal) in multiresolution form. The wavelet transform was used by many researchers for texture analysis [1] and image retrieval [2]. The main problem with wavelet analysis is that it is not rotation or translation invariant. To solve this problem, Freeman et. al [3] introduced a structure called steerable wavelets or steerable pyramids. This transform is overcomplete with higehr complexity than the ordinary wavelet transform but it is rotation and translation invariant.

The purpose of the proposed master thesis is to try to extract features that are invariant with respect to rotation and translation from the steerable wavelets and to use the features for image retrieval.

Bibliography

1

S. Livens, P. Scheunders, G. Van de Wouwer, D. Van Dyck, "Wavelets for Texture Analysis ", In Proc. IEE Conf. on Image Proc. and Its Applications, July,1997.

2

Charles E. Jacobs, Adam Finkelstein, David H. Salesin, "Fast Multiresolution Image Querying ", Proceedings of the 22nd annual conference on Computer graphics and interactive techniques, 1995.

3

William T. Freeman, Edward H. Adelson, "The Design and Use of Steerable Filters ", IEEE Trans. Pattern Analysis and Machine Intelligence, vol.13, no.9, pp. 891-906, 1991.

Candidates: Students in electrical engineering, physics, computer science or mathematics. Knowledge of a high-level programming language is desirable.

Please contact:

Betreuername: Alaa Halawani
Raum: 01-023
Telefon: 203-8271
E-Mail: halawani@informatik.uni-freiburg.de

March 2005

Gerd Brunner 2005-15-03