Conference proceeding
Group theoretic approach to recovery of omnidirectional images
Proceedings of the 8th Joint Conference on Information Sciences, Vols 1-3, pp.554-557
01 Jan 2005
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Omnidirectional images have become important in the field of computer vision as they offer larger and more complex views of the world. Image processing of such images requires an approach that accounts for symmetries of the image and the possible transformations and deformations of objects within the image. Group representation theory already provides the framework for a general Fourier transform and its inverse over the 2D Euclidean motion group SE(2). We found that the operator-valued Fourier transform on the motion group reduces for a two-dimensional image to a standard Fourier transform in polar form. This observation allows bypassing the complicated inversion formula for non-commutative groups. In this paper, a numerical implementation of the SE(2) Fourier transform is used to recover omnidirectional images from the polar frequency domain. Further, we show that a simple splitting of the transform before inversion significantly improves image reconstruction.
Metrics
4 Record Views
Details
- Title
- Group theoretic approach to recovery of omnidirectional images
- Creators
- Amal AafifRobert Boyer
- Contributors
- U Chakraborty (Editor)S Blair (Editor)
- Publication Details
- Proceedings of the 8th Joint Conference on Information Sciences, Vols 1-3, pp.554-557
- Conference
- 8th Joint Conference on Information Sciences, 8th (Salt Lake City, Utah, United States, 21 Jul 2005 - 26 Jul 2005)
- Publisher
- Joint Conference Information Sciences
- Number of pages
- 4
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Mathematics
- Identifiers
- 991019170406304721
UN Sustainable Development Goals (SDGs)
This output has contributed to the advancement of the following goals:
InCites Highlights
These are selected metrics from InCites Benchmarking & Analytics tool, related to this output
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Information Systems
- Computer Science, Interdisciplinary Applications