Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information

Chul Hee Lee, Sungwook Youn, Taeuk Jeong, Eunjae Lee, Joan Serra-Sagristà

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

It has been shown that image compression based on principal component analysis (PCA) provides good compression efficiency for hyperspectral images. However, PCA might fail to capture all the discriminant information of hyperspectral images, since features that are important for classification tasks may not be high in signal energy. To deal with this problem, we propose a hybrid compression method for hyperspectral images with pre-encoding discriminant information. A feature extraction method is first applied to the original images, producing a set of feature vectors that are used to generate feature images and then residual images by subtracting the feature-reconstructed images from the original ones. Both feature images and residual images are compressed and transmitted. Experiments on data from the Airborne Visible/Infrared Imaging Spectrometer sensor indicate that the proposed method provides better compression efficiency with improved classification accuracy than conventional compression methods.

Original languageEnglish
Article number7063895
Pages (from-to)1491-1495
Number of pages5
JournalIEEE Geoscience and Remote Sensing Letters
Volume12
Issue number7
DOIs
Publication statusPublished - 2015 Jul 1

Fingerprint

Principal component analysis
principal component analysis
compression
Infrared imaging
Image compression
Spectrometers
Feature extraction
Sensors
Experiments
AVIRIS
extraction method
sensor

All Science Journal Classification (ASJC) codes

  • Geotechnical Engineering and Engineering Geology
  • Electrical and Electronic Engineering

Cite this

Lee, Chul Hee ; Youn, Sungwook ; Jeong, Taeuk ; Lee, Eunjae ; Serra-Sagristà, Joan. / Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information. In: IEEE Geoscience and Remote Sensing Letters. 2015 ; Vol. 12, No. 7. pp. 1491-1495.
@article{483423f0d60a467b8753679ef33d626b,
title = "Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information",
abstract = "It has been shown that image compression based on principal component analysis (PCA) provides good compression efficiency for hyperspectral images. However, PCA might fail to capture all the discriminant information of hyperspectral images, since features that are important for classification tasks may not be high in signal energy. To deal with this problem, we propose a hybrid compression method for hyperspectral images with pre-encoding discriminant information. A feature extraction method is first applied to the original images, producing a set of feature vectors that are used to generate feature images and then residual images by subtracting the feature-reconstructed images from the original ones. Both feature images and residual images are compressed and transmitted. Experiments on data from the Airborne Visible/Infrared Imaging Spectrometer sensor indicate that the proposed method provides better compression efficiency with improved classification accuracy than conventional compression methods.",
author = "Lee, {Chul Hee} and Sungwook Youn and Taeuk Jeong and Eunjae Lee and Joan Serra-Sagrist{\`a}",
year = "2015",
month = "7",
day = "1",
doi = "10.1109/LGRS.2015.2409897",
language = "English",
volume = "12",
pages = "1491--1495",
journal = "IEEE Geoscience and Remote Sensing Letters",
issn = "1545-598X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information. / Lee, Chul Hee; Youn, Sungwook; Jeong, Taeuk; Lee, Eunjae; Serra-Sagristà, Joan.

In: IEEE Geoscience and Remote Sensing Letters, Vol. 12, No. 7, 7063895, 01.07.2015, p. 1491-1495.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hybrid Compression of Hyperspectral Images Based on PCA With Pre-Encoding Discriminant Information

AU - Lee, Chul Hee

AU - Youn, Sungwook

AU - Jeong, Taeuk

AU - Lee, Eunjae

AU - Serra-Sagristà, Joan

PY - 2015/7/1

Y1 - 2015/7/1

N2 - It has been shown that image compression based on principal component analysis (PCA) provides good compression efficiency for hyperspectral images. However, PCA might fail to capture all the discriminant information of hyperspectral images, since features that are important for classification tasks may not be high in signal energy. To deal with this problem, we propose a hybrid compression method for hyperspectral images with pre-encoding discriminant information. A feature extraction method is first applied to the original images, producing a set of feature vectors that are used to generate feature images and then residual images by subtracting the feature-reconstructed images from the original ones. Both feature images and residual images are compressed and transmitted. Experiments on data from the Airborne Visible/Infrared Imaging Spectrometer sensor indicate that the proposed method provides better compression efficiency with improved classification accuracy than conventional compression methods.

AB - It has been shown that image compression based on principal component analysis (PCA) provides good compression efficiency for hyperspectral images. However, PCA might fail to capture all the discriminant information of hyperspectral images, since features that are important for classification tasks may not be high in signal energy. To deal with this problem, we propose a hybrid compression method for hyperspectral images with pre-encoding discriminant information. A feature extraction method is first applied to the original images, producing a set of feature vectors that are used to generate feature images and then residual images by subtracting the feature-reconstructed images from the original ones. Both feature images and residual images are compressed and transmitted. Experiments on data from the Airborne Visible/Infrared Imaging Spectrometer sensor indicate that the proposed method provides better compression efficiency with improved classification accuracy than conventional compression methods.

UR - http://www.scopus.com/inward/record.url?scp=85027950675&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027950675&partnerID=8YFLogxK

U2 - 10.1109/LGRS.2015.2409897

DO - 10.1109/LGRS.2015.2409897

M3 - Article

AN - SCOPUS:85027950675

VL - 12

SP - 1491

EP - 1495

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

SN - 1545-598X

IS - 7

M1 - 7063895

ER -