Extended morphological profiles using auto-associative neural networks for hyperspectral data classification

Giorgio Licciardi; Prashanth Reddy Marpu; Jon Atli Benediktsson; Jocelyn Chanussot

doi:10.1109/WHISPERS.2011.6080867

Extended morphological profiles using auto-associative neural networks for hyperspectral data classification

Giorgio Licciardi
, Prashanth Reddy Marpu
, Jon Atli Benediktsson
, Jocelyn Chanussot

University of Iceland

University of Iceland

Research output: Contribution to journal › Conference article › peer-review

Abstract

Recently, morphological profiles have be observed as good tools to fuse spectral and spatial information to produce better classification results. In general, the profiles are built with the features derived using the principal component analysis (PCA). Auto-associative neural network (AANN), which can be seen as an implementation of non-linear PCA is used for unsupervised feature reduction of hyperspectral data. In this paper, we investigate the suitability of the features derived using AANN to build extended morphological profiles for hyperspectral data classification.

Original language	English
Article number	6080867
Journal	Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
DOIs	https://doi.org/10.1109/WHISPERS.2011.6080867
Publication status	Published - 2011
Event	3rd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2011 - Lisbon, Portugal Duration: 6 Jun 2011 → 9 Jun 2011

Other keywords

Morphological profiles
auto-associative neural networks
classification
feature reduction

Access to Document

10.1109/WHISPERS.2011.6080867

Cite this

@article{2301939a20c8430d9035e38fdf9b512c,

title = "Extended morphological profiles using auto-associative neural networks for hyperspectral data classification",

abstract = "Recently, morphological profiles have be observed as good tools to fuse spectral and spatial information to produce better classification results. In general, the profiles are built with the features derived using the principal component analysis (PCA). Auto-associative neural network (AANN), which can be seen as an implementation of non-linear PCA is used for unsupervised feature reduction of hyperspectral data. In this paper, we investigate the suitability of the features derived using AANN to build extended morphological profiles for hyperspectral data classification.",

keywords = "Morphological profiles, auto-associative neural networks, classification, feature reduction",

author = "Giorgio Licciardi and Marpu, \{Prashanth Reddy\} and Benediktsson, \{Jon Atli\} and Jocelyn Chanussot",

year = "2011",

doi = "10.1109/WHISPERS.2011.6080867",

language = "English",

journal = "Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing",

issn = "2158-6276",

publisher = "IEEE Computer Society",

note = "3rd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2011 ; Conference date: 06-06-2011 Through 09-06-2011",

}

TY - JOUR

T1 - Extended morphological profiles using auto-associative neural networks for hyperspectral data classification

AU - Licciardi, Giorgio

AU - Marpu, Prashanth Reddy

AU - Benediktsson, Jon Atli

AU - Chanussot, Jocelyn

PY - 2011

Y1 - 2011

N2 - Recently, morphological profiles have be observed as good tools to fuse spectral and spatial information to produce better classification results. In general, the profiles are built with the features derived using the principal component analysis (PCA). Auto-associative neural network (AANN), which can be seen as an implementation of non-linear PCA is used for unsupervised feature reduction of hyperspectral data. In this paper, we investigate the suitability of the features derived using AANN to build extended morphological profiles for hyperspectral data classification.

AB - Recently, morphological profiles have be observed as good tools to fuse spectral and spatial information to produce better classification results. In general, the profiles are built with the features derived using the principal component analysis (PCA). Auto-associative neural network (AANN), which can be seen as an implementation of non-linear PCA is used for unsupervised feature reduction of hyperspectral data. In this paper, we investigate the suitability of the features derived using AANN to build extended morphological profiles for hyperspectral data classification.

KW - Morphological profiles

KW - auto-associative neural networks

KW - classification

KW - feature reduction

UR - https://www.scopus.com/pages/publications/84255182525

U2 - 10.1109/WHISPERS.2011.6080867

DO - 10.1109/WHISPERS.2011.6080867

M3 - Conference article

SN - 2158-6276

JO - Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing

JF - Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing

M1 - 6080867

T2 - 3rd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2011

Y2 - 6 June 2011 through 9 June 2011

ER -

Extended morphological profiles using auto-associative neural networks for hyperspectral data classification

Abstract

Other keywords

Access to Document

Fingerprint

Cite this