An improved semi-supervised local discriminant analysis for feature extraction of hyperspectal image

Renbo Luo; Wenzhi Liao; Wilfried Philips; Youguo Pi

doi:10.1109/JURSE.2015.7120508

An improved semi-supervised local discriminant analysis for feature extraction of hyperspectal image

Renbo Luo, Wenzhi Liao, Wilfried Philips, Youguo Pi

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

8 Citations (Scopus)

Abstract

We propose an improved semi-supervised local discriminant analysis (ISELD) for feature extraction of hyperspectral image in this paper. The proposed ISELD method aims to find a projection which can preserve local neighborhood information and maximize the class discrimination of the data. Compared to the previous SELD, the proposed ISELD better models the correlation of labeled and unlabeled samples. Experimental results on an ROSIS urban hyperspectral image are encouraging. Compared to some recent feature extraction methods, our approach has more than 2% improvements as the training sample size changes.

Original language	English
Number of pages	4
DOIs	https://doi.org/10.1109/JURSE.2015.7120508
Publication status	Published - 11 Jun 2015
Event	2015 JOINT URBAN REMOTE SENSING EVENT (JURSE) - Lausanne, Switzerland Duration: 30 Mar 2015 → 1 Apr 2015

Conference

Conference	2015 JOINT URBAN REMOTE SENSING EVENT (JURSE)
Abbreviated title	JURSE 2015
Country/Territory	Switzerland
City	Lausanne
Period	30/03/15 → 1/04/15

Keywords

feature extraction
soil
asphalt
principal component analysis
hyperspectral imaging
image recognition
learning (artificial intelligence)
remote sensing

Access to Document

10.1109/JURSE.2015.7120508

Cite this

@conference{6a556f074f984d5eb919ca6527b8a7ab,

title = "An improved semi-supervised local discriminant analysis for feature extraction of hyperspectal image",

abstract = "We propose an improved semi-supervised local discriminant analysis (ISELD) for feature extraction of hyperspectral image in this paper. The proposed ISELD method aims to find a projection which can preserve local neighborhood information and maximize the class discrimination of the data. Compared to the previous SELD, the proposed ISELD better models the correlation of labeled and unlabeled samples. Experimental results on an ROSIS urban hyperspectral image are encouraging. Compared to some recent feature extraction methods, our approach has more than 2% improvements as the training sample size changes.",

keywords = "feature extraction, soil, asphalt, principal component analysis, hyperspectral imaging, image recognition, learning (artificial intelligence), remote sensing",

author = "Renbo Luo and Wenzhi Liao and Wilfried Philips and Youguo Pi",

year = "2015",

month = jun,

day = "11",

doi = "10.1109/JURSE.2015.7120508",

language = "English",

note = "2015 JOINT URBAN REMOTE SENSING EVENT (JURSE), JURSE 2015 ; Conference date: 30-03-2015 Through 01-04-2015",

}

TY - CONF

T1 - An improved semi-supervised local discriminant analysis for feature extraction of hyperspectal image

AU - Luo, Renbo

AU - Liao, Wenzhi

AU - Philips, Wilfried

AU - Pi, Youguo

PY - 2015/6/11

Y1 - 2015/6/11

N2 - We propose an improved semi-supervised local discriminant analysis (ISELD) for feature extraction of hyperspectral image in this paper. The proposed ISELD method aims to find a projection which can preserve local neighborhood information and maximize the class discrimination of the data. Compared to the previous SELD, the proposed ISELD better models the correlation of labeled and unlabeled samples. Experimental results on an ROSIS urban hyperspectral image are encouraging. Compared to some recent feature extraction methods, our approach has more than 2% improvements as the training sample size changes.

AB - We propose an improved semi-supervised local discriminant analysis (ISELD) for feature extraction of hyperspectral image in this paper. The proposed ISELD method aims to find a projection which can preserve local neighborhood information and maximize the class discrimination of the data. Compared to the previous SELD, the proposed ISELD better models the correlation of labeled and unlabeled samples. Experimental results on an ROSIS urban hyperspectral image are encouraging. Compared to some recent feature extraction methods, our approach has more than 2% improvements as the training sample size changes.

KW - feature extraction

KW - soil

KW - asphalt

KW - principal component analysis

KW - hyperspectral imaging

KW - image recognition

KW - learning (artificial intelligence)

KW - remote sensing

UR - http://hdl.handle.net/1854/LU-5898049

U2 - 10.1109/JURSE.2015.7120508

DO - 10.1109/JURSE.2015.7120508

M3 - Paper

T2 - 2015 JOINT URBAN REMOTE SENSING EVENT (JURSE)

Y2 - 30 March 2015 through 1 April 2015

ER -

An improved semi-supervised local discriminant analysis for feature extraction of hyperspectal image

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this