Hyperspectral Image Classification Using Compressive Sampling Measurements

DOI

10.2991/jimec-17.2017.89How to use a DOI?

Keywords

Compressive Sensing, Classification, Hyperspectral Image, OMP

Abstract

In this paper, we develop a new approach for hyperspectral image classification directly from the compressive sensing measurements without reconstructing the original hyperspectral image first. The proposed method is based on the fact that each pixel in the hyperspectral image lies in a low-dimensional subspace, and thus it can be represented as a sparse linear combination of vectors in a dictionary obtained from training samples. In compressive sensing theory, with the sparsity prior, we can reconstruct the original signal from the random sampling measurements using appropriate algorithms. And finally the recovered sparse vector is used to determine the class label of the test pixel by the nearest neighbor classifier. The proposed method can fulfil the classification task and reconstruction at the same time.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chen Xinmeng
AU  - Li Yuting
AU  - Liu Jiying
AU  - Zhu Jubo
PY  - 2017/10
DA  - 2017/10
TI  - Hyperspectral Image Classification Using Compressive Sampling Measurements
BT  - Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)
PB  - Atlantis Press
SP  - 406
EP  - 409
SN  - 2352-538X
UR  - https://doi.org/10.2991/jimec-17.2017.89
DO  - 10.2991/jimec-17.2017.89
ID  - Xinmeng2017/10
ER  -