Two-Stream Deep Architecture for Hyperspectral Image Classification

Most traditional approaches classify hyperspectral image (HSI) pixels relying only on the spectral values of the input channels. However, the spatial context around a pixel is also very important and can enhance the classification performance. In order to effectively exploit and fuse both the spatia...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on geoscience and remote sensing Jg. 56; H. 4; S. 2349 - 2361
Hauptverfasser: Hao, Siyuan, Wang, Wei, Ye, Yuanxin, Nie, Tingyuan, Bruzzone, Lorenzo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.04.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Architecture
Artificial neural networks
Class-specific fusion
Classification
convolutional neural networks (CNNs)
deep learning
Feature extraction
hyperspectral image (HSI) classification
Hyperspectral imaging
Image classification
Machine learning
Neural networks
Noise reduction
Pixels
remote sensing
Rivers
Spectra
stacked denoising autoencoder (SdAE)
State of the art
Training
two-stream architecture
Weight
ISSN:0196-2892, 1558-0644
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Most traditional approaches classify hyperspectral image (HSI) pixels relying only on the spectral values of the input channels. However, the spatial context around a pixel is also very important and can enhance the classification performance. In order to effectively exploit and fuse both the spatial context and spectral structure, we propose a novel two-stream deep architecture for HSI classification. The proposed method consists of a two-stream architecture and a novel fusion scheme. In the two-stream architecture, one stream employs the stacked denoising autoencoder to encode the spectral values of each input pixel, and the other stream takes as input the corresponding image patch and deep convolutional neural networks are employed to process the image patch. In the fusion scheme, the prediction probabilities from two streams are fused by adaptive class-specific weights, which can be obtained by a fully connected layer. Finally, a weight regularizer is added to the loss function to alleviate the overfitting of the class-specific fusion weights. Experimental results on real HSIs demonstrate that the proposed two-stream deep architecture can achieve competitive performance compared with the state-of-the-art methods.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2017.2778343