Mitigating discontinuities in segmented Karhunen-Loeve Transforms

Persistent Link:
http://hdl.handle.net/10150/623189
Title:
Mitigating discontinuities in segmented Karhunen-Loeve Transforms
Author:
Stadnicka, Monika; Blanes, Ian; Serra-Sagrista, Joan; Marcellin, Michael W.
Affiliation:
Univ Arizona, Dept Elect & Comp Engn
Issue Date:
2016-09
Publisher:
IEEE
Citation:
M. Stadnicka, I. Blanes, J. Serra-Sagristà and M. W. Marcellin, "Mitigating discontinuities in segmented Karhunen-Loeve Transforms," 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, 2016, pp. 2211-2215. doi: 10.1109/ICIP.2016.7532751
Journal:
2016 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)
Rights:
© 2016, IEEE
Collection Information:
This item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at repository@u.library.arizona.edu.
Abstract:
The Karhunen-Loeve Transform (KLT) is a popular transform used in multiple image processing scenarios. Sometimes, the application of the KLT is not carried out as a single transform over an entire image Rather, the image is divided into smaller spatial regions (segments), each of which is transformed by a smaller dimensional KLT. Such a situation may penalize the transform efficiency. An improvement for the segmented KLT, aiming at mitigating discontinuities arising on the edge of adjacent regions, is proposed in this paper. In the case of moderately varying image regions, discontinuities occur as the consequence of disregarded similarity between transform domains, as the order and sign of eigenvectors in the transform matrices are mismatched. In the proposed method, the KLT is adjusted to guarantee the best achievable similarity via the optimal assignment and sign correspondence for eigenvectors. Experimental results indicate that the proposed transform improves the similarity between transform domains, and reduces RMSE on the edge of adjacent regions. In consequence, images processed by the adjusted KLT present better cohesion and continuity between independently transformed regions.
ISSN:
1522-4880
DOI:
10.1109/ICIP.2016.7532751
Keywords:
Karhunen-Loeve Transform; KLT; Segmentation; Discontinuities; Hyperspectral Images
Version:
Final accepted manuscript
Sponsors:
This work was supported in part by FEDER, the Spanish Government (MINECO), and the Catalan Government, under grants TIN2015-71126-R and 2014SGR-691.
Additional Links:
http://ieeexplore.ieee.org/document/7532751/

Full metadata record

DC FieldValue Language
dc.contributor.authorStadnicka, Monikaen
dc.contributor.authorBlanes, Ianen
dc.contributor.authorSerra-Sagrista, Joanen
dc.contributor.authorMarcellin, Michael W.en
dc.date.accessioned2017-04-21T00:57:43Z-
dc.date.available2017-04-21T00:57:43Z-
dc.date.issued2016-09-
dc.identifier.citationM. Stadnicka, I. Blanes, J. Serra-Sagristà and M. W. Marcellin, "Mitigating discontinuities in segmented Karhunen-Loeve Transforms," 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, 2016, pp. 2211-2215. doi: 10.1109/ICIP.2016.7532751en
dc.identifier.issn1522-4880-
dc.identifier.doi10.1109/ICIP.2016.7532751-
dc.identifier.urihttp://hdl.handle.net/10150/623189-
dc.description.abstractThe Karhunen-Loeve Transform (KLT) is a popular transform used in multiple image processing scenarios. Sometimes, the application of the KLT is not carried out as a single transform over an entire image Rather, the image is divided into smaller spatial regions (segments), each of which is transformed by a smaller dimensional KLT. Such a situation may penalize the transform efficiency. An improvement for the segmented KLT, aiming at mitigating discontinuities arising on the edge of adjacent regions, is proposed in this paper. In the case of moderately varying image regions, discontinuities occur as the consequence of disregarded similarity between transform domains, as the order and sign of eigenvectors in the transform matrices are mismatched. In the proposed method, the KLT is adjusted to guarantee the best achievable similarity via the optimal assignment and sign correspondence for eigenvectors. Experimental results indicate that the proposed transform improves the similarity between transform domains, and reduces RMSE on the edge of adjacent regions. In consequence, images processed by the adjusted KLT present better cohesion and continuity between independently transformed regions.en
dc.description.sponsorshipThis work was supported in part by FEDER, the Spanish Government (MINECO), and the Catalan Government, under grants TIN2015-71126-R and 2014SGR-691.en
dc.language.isoenen
dc.publisherIEEEen
dc.relation.urlhttp://ieeexplore.ieee.org/document/7532751/en
dc.rights© 2016, IEEEen
dc.subjectKarhunen-Loeve Transformen
dc.subjectKLTen
dc.subjectSegmentationen
dc.subjectDiscontinuitiesen
dc.subjectHyperspectral Imagesen
dc.titleMitigating discontinuities in segmented Karhunen-Loeve Transformsen
dc.typeProceedingsen
dc.contributor.departmentUniv Arizona, Dept Elect & Comp Engnen
dc.identifier.journal2016 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)en
dc.description.collectioninformationThis item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at repository@u.library.arizona.edu.en
dc.eprint.versionFinal accepted manuscripten
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.