Estimation of Multiple Orientations in Multi-Dimensional Signals

Cicero Mota, Til Aach, Ingo Stuke, Erhardt Barth

Abstract

We present a solution to the general problem of estimating multiple orientations in multi-dimensional signals. The solution is divided in a linear part that provides the mixed-orientation space (MOS), and a non-linear part that gives the actual orientation spaces. We show that the angle between two overlaid orientations is an invariant that can be derived from the MOS without solving the non-linear part, and that all other invariants are generated by this angle. Results obtained for synthetic images illustrate that the above invariant is a useful image feature for various applications such as pattern recognition and texture segmentation.

Original languageEnglish
Title of host publication 2004 International Conference on Image Processing, 2004. ICIP '04.
Number of pages4
PublisherIEEE
Publication date01.12.2004
Pages2665-2668
ISBN (Print)0-7803-8554-3
DOIs
Publication statusPublished - 01.12.2004
Event2004 International Conference on Image Processing
- Singapore, Singapore
Duration: 18.10.200421.10.2004
Conference number: 65024

Fingerprint

Dive into the research topics of 'Estimation of Multiple Orientations in Multi-Dimensional Signals'. Together they form a unique fingerprint.

Cite this