Paper Info
Title
Semi-Supervised Deep Learning Classification for Hyperspectral Image Based on Dual-Strategy Sample Selection.
Abstract
This paper studies the classification problem of hyperspectral image (HSI). Inspired by the great success of deep neural networks in Artificial Intelligence (Al), researchers have proposed different deep learning based algorithms to improve the performance of hyperspectral classification. However, deep learning based algorithms always require a large-scale annotated dataset to provide sufficient training. To address this problem, we propose a semi-supervised deep learning framework based on the residual networks (ResNets), which use very limited labeled data supplemented by abundant unlabeled data. The core of our framework is a novel dual-strategy sample selection co-training algorithm, which can successfully guide ResNets to learn from the unlabeled data by making full use of the complementary cues of the spectral and spatial features in HSI classification. Experiments on the benchmark HSI dataset and real HSI dataset demonstrate that, with a small number of training data, our approach achieves competitive performance with maximum improvement of 41% (compare with traditional convolutional neural network (CNN) with 5 initial training samples per class on Indian Pines dataset) for HSI classification as compared with the results from those state-of-the-art supervised and semi-supervised methods.
Year
DOI
Venue
2018
10.3390/rs10040574
REMOTE SENSING
Keywords
Field
DocType
hyperspectral image classification,deep learning,residual networks,co-training,sample selection
Computer vision,Residual,Pattern recognition,Convolutional neural network,Co-training,Image based,Hyperspectral imaging,Artificial intelligence,Labeled data,Deep learning,Sample selection,Geology
Journal
Volume
Issue
Citations 
10
4
2
PageRank 
References 
Authors
0.37
10
4
Name
Order
Citations
PageRank
Bei Fang150.73
Ying Li213021.36
Haokui Zhang3715.71
Jonathan Cheung-Wai Chan415518.46