TY - JOUR
T1 - Spatial-spectral classification of hyperspectral images
T2 - a deep learning framework with Markov random fields based modeling
AU - Qing, Chunmei
AU - Ruan, Jiawei
AU - Xu, Xiangmin
AU - Ren, Jinchang
AU - Zabalza, Jaime
N1 - This paper is a postprint of a paper submitted to and accepted for publication in IET Image Processing and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at IET Digital Library.
PY - 2019/2/25
Y1 - 2019/2/25
N2 - For spatial-spectral classification of hyperspectral images (HSI), a deep learning framework is proposed in this paper, which consists of convolutional neural networks (CNN) and Markov random fields (MRF). Firstly, a CNN model to learn the deep spectral feature from the HSI is built and the class posterior probability distribution is estimated. The CNN with a dropout layer can relieve the overfitting in classification. The CNN is utilized as a pixel-classifier, so it only works in the spectral domain. Then, the spatial information will be encoded by MRF-based multilevel logistic (MLL) prior for regularizing the classification. To derive the correlation of both spectral and spatial features for improving algorithm performance, the marginal probability distribution in HSI is learned using MRF-based loopy belief propagation (LBP). In comparison with several state-of-the-art approaches for data classification on 3 publicly available HSI datasets, experimental results have demonstrated the superior performance of the proposed methodology.
AB - For spatial-spectral classification of hyperspectral images (HSI), a deep learning framework is proposed in this paper, which consists of convolutional neural networks (CNN) and Markov random fields (MRF). Firstly, a CNN model to learn the deep spectral feature from the HSI is built and the class posterior probability distribution is estimated. The CNN with a dropout layer can relieve the overfitting in classification. The CNN is utilized as a pixel-classifier, so it only works in the spectral domain. Then, the spatial information will be encoded by MRF-based multilevel logistic (MLL) prior for regularizing the classification. To derive the correlation of both spectral and spatial features for improving algorithm performance, the marginal probability distribution in HSI is learned using MRF-based loopy belief propagation (LBP). In comparison with several state-of-the-art approaches for data classification on 3 publicly available HSI datasets, experimental results have demonstrated the superior performance of the proposed methodology.
KW - hyperspectral image
KW - spatial-spectral classification
KW - convolutional neural networks
KW - Markov random fields
KW - loopy belief propagation
UR - http://digital-library.theiet.org/content/journals/iet-ipr
U2 - 10.1049/iet-ipr.2018.5727
DO - 10.1049/iet-ipr.2018.5727
M3 - Special issue
VL - 13
SP - 235
EP - 245
JO - IET Image Processing
JF - IET Image Processing
SN - 1751-9659
IS - 2
ER -