DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Unsupervised Spectral-Spatial Feature Learning via Deep Residual Conv-Deconv Network for Hyperspectral Image Classification

Mou, Lichao and Ghamisi, Pedram and Zhu, Xiaoxiang (2018) Unsupervised Spectral-Spatial Feature Learning via Deep Residual Conv-Deconv Network for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 56 (1), pp. 391-406. IEEE - Institute of Electrical and Electronics Engineers. doi: 10.1109/TGRS.2017.2748160. ISSN 0196-2892.

[img] PDF

Official URL: http://ieeexplore.ieee.org/document/8082108/


Supervised approaches classify input data using a set of representative samples for each class, known as Training samples. The collection of such samples is expensive and time demanding. Hence, unsupervised feature learning, which has a quick access to arbitrary amounts of unlabeled data, is conceptually of high interest. In this paper, we propose a novel network architecture, fully Conv–Deconv network, for unsupervised spectral–spatial feature learning of hyperspectral images, which is able to be trained in an end-to-end manner. Specifically, our network is based on the so-called encoder– decoder paradigm, i.e., the input 3-D hyperspectral patch is first transformed into a typically lower dimensional space via a convolutional subnetwork (encoder), and then expanded to reproduce the initial data by a deconvolutional subnetwork (decoder). However, during the experiment, we found that such a network is not easy to be optimized. To address this problem, we refine the proposed network architecture by incorporating: 1) residual learning and 2) a new unpooling operation that can use memorized max-pooling indexes. Moreover, to understand the “black box,” we make an in-depth study of the learned Feature maps in the experimental analysis. A very interesting discovery is that some specific “neurons” in the first residual block of the proposed network own good description power for semantic visual patterns in the object level, which provide an opportunity to achieve “free” object detection. This paper, for the first time in the remote sensing community, proposes an end-to-end fully Conv–Deconv network for unsupervised spectral–spatial feature learning. Moreover, this paper also introduces an in-depth investigation of learned features. Experimental results on two widely used hyperspectral data, Indian Pines and Pavia University, demonstrate competitive performance obtained by the proposed methodology compared with other studied approaches.

Item URL in elib:https://elib.dlr.de/114209/
Document Type:Article
Title:Unsupervised Spectral-Spatial Feature Learning via Deep Residual Conv-Deconv Network for Hyperspectral Image Classification
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Mou, Lichaolichao.mou (at) dlr.deUNSPECIFIED
Ghamisi, Pedrampedram.ghamisi (at) dlr.deUNSPECIFIED
Zhu, Xiaoxiangxiao.zhu (at) dlr.deUNSPECIFIED
Date:January 2018
Journal or Publication Title:IEEE Transactions on Geoscience and Remote Sensing
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:Yes
DOI :10.1109/TGRS.2017.2748160
Page Range:pp. 391-406
Publisher:IEEE - Institute of Electrical and Electronics Engineers
Keywords:Convolutional network, deconvolutional network, hyperspectral image classification, residual learning, unsupervised spectral-spatial feature learning
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Vorhaben hochauflösende Fernerkundungsverfahren (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > SAR Signal Processing
Deposited By: Mou, LiChao
Deposited On:19 Sep 2017 11:42
Last Modified:31 Jul 2019 20:11

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.