The University of Southampton
University of Southampton Institutional Repository

Image reconstruction from local binary patterns

Image reconstruction from local binary patterns
Image reconstruction from local binary patterns
Reconstruction of an image from its LBP codes can aid understanding of the information contained within the codes by comparing the reconstructed image to the original. We are the first to show that the LBP process can be inverted and present a novel algorithm to perform the reconstruction, resulting in an approximation of the original image that is both visually appealing and completely matches the LBP codes of the original. The algorithm calculates the minimum contrast between two pixels; reconstructing some of the contrast information thought lost in the LBP process. Tests on the algorithm have been conducted on images from the Brodatz database and Berkeley Segmentation Dataset which show an image visually similar to the original with perfect texture construction. The reconstructed images also remove the effects of illumination from the images, suggesting future investigation into the possibility of image brightness normalisation. Additionally, since the reconstructed image provides the same LBP codes as the original, the susceptibility to spoofing of systems using LBP feature vectors has been identified.
118-123
Waller, Ben
a802857c-78ee-4290-a400-20d9ef2b0f5e
Nixon, Mark S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Carter, John N.
e05be2f9-991d-4476-bb50-ae91606389da
Waller, Ben
a802857c-78ee-4290-a400-20d9ef2b0f5e
Nixon, Mark S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Carter, John N.
e05be2f9-991d-4476-bb50-ae91606389da

Waller, Ben, Nixon, Mark S. and Carter, John N. (2013) Image reconstruction from local binary patterns. IEEE International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), Japan. 02 - 05 Dec 2013. pp. 118-123 . (doi:10.1109/SITIS.2013.30).

Record type: Conference or Workshop Item (Paper)

Abstract

Reconstruction of an image from its LBP codes can aid understanding of the information contained within the codes by comparing the reconstructed image to the original. We are the first to show that the LBP process can be inverted and present a novel algorithm to perform the reconstruction, resulting in an approximation of the original image that is both visually appealing and completely matches the LBP codes of the original. The algorithm calculates the minimum contrast between two pixels; reconstructing some of the contrast information thought lost in the LBP process. Tests on the algorithm have been conducted on images from the Brodatz database and Berkeley Segmentation Dataset which show an image visually similar to the original with perfect texture construction. The reconstructed images also remove the effects of illumination from the images, suggesting future investigation into the possibility of image brightness normalisation. Additionally, since the reconstructed image provides the same LBP codes as the original, the susceptibility to spoofing of systems using LBP feature vectors has been identified.

Text
waller sitis 2013.pdf - Version of Record
Restricted to Repository staff only
Request a copy

More information

Published date: December 2013
Venue - Dates: IEEE International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), Japan, 2013-12-02 - 2013-12-05
Organisations: Vision, Learning and Control

Identifiers

Local EPrints ID: 363304
URI: http://eprints.soton.ac.uk/id/eprint/363304
PURE UUID: daf6dd85-addf-441b-a204-3321e9eaa5b0
ORCID for Mark S. Nixon: ORCID iD orcid.org/0000-0002-9174-5934

Catalogue record

Date deposited: 20 Mar 2014 16:24
Last modified: 29 Oct 2019 02:11

Export record

Altmetrics

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×