Quantization of hyperspectral image manifold using probabilistic distances

Gianni Franchi; Jesús Angulo

doi:10.1007/978-3-319-25040-3_44

Quantization of hyperspectral image manifold using probabilistic distances

Gianni Franchi, Jesús Angulo

Mines ParisTech

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

A technique of spatial-spectral quantization of hyperspectral images is introduced. Thus a quantized hyperspectral image is just summarized by K spectra which represent the spatial and spectral structures of the image. The proposed technique is based on α-connected components on a region adjacency graph. The main ingredient is a dissimilarity metric. In order to choose the metric that best fit the hyperspectral data manifold, a comparison of different probabilistic dissimilarity measures is achieved.

Original language	English
Title of host publication	Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings
Editors	Frank Nielsen, Frank Nielsen, Frank Nielsen, Frederic Barbaresco, Frederic Barbaresco, Frank Nielsen
Publisher	Springer Verlag
Pages	406-414
Number of pages	9
ISBN (Print)	9783319250397, 9783319250397
DOIs	https://doi.org/10.1007/978-3-319-25040-3_44
Publication status	Published - 1 Jan 2015
Externally published	Yes
Event	2nd International Conference on Geometric Science of Information, GSI 2015 - Palaiseau, France Duration: 28 Oct 2015 → 30 Oct 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9389
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	2nd International Conference on Geometric Science of Information, GSI 2015
Country/Territory	France
City	Palaiseau
Period	28/10/15 → 30/10/15

Keywords

Hyperspectral images
Information geometry
Mathematical morphology
Probabilistic distances
Quantization

Access to Document

10.1007/978-3-319-25040-3_44

Cite this

Franchi, G., & Angulo, J. (2015). Quantization of hyperspectral image manifold using probabilistic distances. In F. Nielsen, F. Nielsen, F. Nielsen, F. Barbaresco, F. Barbaresco, & F. Nielsen (Eds.), Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings (pp. 406-414). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9389). Springer Verlag. https://doi.org/10.1007/978-3-319-25040-3_44

Franchi, Gianni ; Angulo, Jesús. / Quantization of hyperspectral image manifold using probabilistic distances. Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. editor / Frank Nielsen ; Frank Nielsen ; Frank Nielsen ; Frederic Barbaresco ; Frederic Barbaresco ; Frank Nielsen. Springer Verlag, 2015. pp. 406-414 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{258e8c30aecf4e43afaad144de37c396,

title = "Quantization of hyperspectral image manifold using probabilistic distances",

abstract = "A technique of spatial-spectral quantization of hyperspectral images is introduced. Thus a quantized hyperspectral image is just summarized by K spectra which represent the spatial and spectral structures of the image. The proposed technique is based on α-connected components on a region adjacency graph. The main ingredient is a dissimilarity metric. In order to choose the metric that best fit the hyperspectral data manifold, a comparison of different probabilistic dissimilarity measures is achieved.",

keywords = "Hyperspectral images, Information geometry, Mathematical morphology, Probabilistic distances, Quantization",

author = "Gianni Franchi and Jes{\'u}s Angulo",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 2nd International Conference on Geometric Science of Information, GSI 2015 ; Conference date: 28-10-2015 Through 30-10-2015",

year = "2015",

month = jan,

day = "1",

doi = "10.1007/978-3-319-25040-3\_44",

language = "English",

isbn = "9783319250397",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "406--414",

editor = "Frank Nielsen and Frank Nielsen and Frank Nielsen and Frederic Barbaresco and Frederic Barbaresco and Frank Nielsen",

booktitle = "Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings",

}

Franchi, G & Angulo, J 2015, Quantization of hyperspectral image manifold using probabilistic distances. in F Nielsen, F Nielsen, F Nielsen, F Barbaresco, F Barbaresco & F Nielsen (eds), Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9389, Springer Verlag, pp. 406-414, 2nd International Conference on Geometric Science of Information, GSI 2015, Palaiseau, France, 28/10/15. https://doi.org/10.1007/978-3-319-25040-3_44

Quantization of hyperspectral image manifold using probabilistic distances. / Franchi, Gianni; Angulo, Jesús.
Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. ed. / Frank Nielsen; Frank Nielsen; Frank Nielsen; Frederic Barbaresco; Frederic Barbaresco; Frank Nielsen. Springer Verlag, 2015. p. 406-414 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9389).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Quantization of hyperspectral image manifold using probabilistic distances

AU - Franchi, Gianni

AU - Angulo, Jesús

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - A technique of spatial-spectral quantization of hyperspectral images is introduced. Thus a quantized hyperspectral image is just summarized by K spectra which represent the spatial and spectral structures of the image. The proposed technique is based on α-connected components on a region adjacency graph. The main ingredient is a dissimilarity metric. In order to choose the metric that best fit the hyperspectral data manifold, a comparison of different probabilistic dissimilarity measures is achieved.

AB - A technique of spatial-spectral quantization of hyperspectral images is introduced. Thus a quantized hyperspectral image is just summarized by K spectra which represent the spatial and spectral structures of the image. The proposed technique is based on α-connected components on a region adjacency graph. The main ingredient is a dissimilarity metric. In order to choose the metric that best fit the hyperspectral data manifold, a comparison of different probabilistic dissimilarity measures is achieved.

KW - Hyperspectral images

KW - Information geometry

KW - Mathematical morphology

KW - Probabilistic distances

KW - Quantization

U2 - 10.1007/978-3-319-25040-3_44

DO - 10.1007/978-3-319-25040-3_44

M3 - Conference contribution

AN - SCOPUS:84950322307

SN - 9783319250397

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 406

EP - 414

BT - Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings

A2 - Nielsen, Frank

A2 - Barbaresco, Frederic

A2 - Nielsen, Frank

PB - Springer Verlag

T2 - 2nd International Conference on Geometric Science of Information, GSI 2015

Y2 - 28 October 2015 through 30 October 2015

ER -

Franchi G, Angulo J. Quantization of hyperspectral image manifold using probabilistic distances. In Nielsen F, Nielsen F, Nielsen F, Barbaresco F, Barbaresco F, Nielsen F, editors, Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Springer Verlag. 2015. p. 406-414. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-25040-3_44

Quantization of hyperspectral image manifold using probabilistic distances

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this