TY - JOUR
T1 - Recursive autoencoders based unsupervised feature learning for hyperspectral image classification
AU - Zhang, Xiangrong
AU - Liang, Yanjie
AU - Li, Chen
AU - Jiao, Licheng
AU - Zhou, Huiyu
PY - 2017/10/11
Y1 - 2017/10/11
N2 - For hyperspectral image (HSI) classification,
it is very important to learn effective features for the discrimination
purpose. Meanwhile, the ability to combine
spectral and spatial information together in a deep level is
also important for feature learning. In this letter, we propose
an unsupervised feature learning method for HSI
classification, which is based on recursive autoencoders
(RAE) network. RAE utilizes the spatial and spectral information
and produces high-level features from the original
data. It learns features from the neighborhood of the
investigated pixel to represent the whole local homogeneous
area of the image. In addition, to obtain more accurate
representation of the investigated pixel, a weighting scheme
is adopted based on the neighboring pixels, where the
weights are determined by the spectral similarity between
the neighboring pixels and the investigated pixel. The effectiveness
of our method is evaluated by the experiments
on two hyperspectral datasets and the results show that our
proposed method has better performance.
AB - For hyperspectral image (HSI) classification,
it is very important to learn effective features for the discrimination
purpose. Meanwhile, the ability to combine
spectral and spatial information together in a deep level is
also important for feature learning. In this letter, we propose
an unsupervised feature learning method for HSI
classification, which is based on recursive autoencoders
(RAE) network. RAE utilizes the spatial and spectral information
and produces high-level features from the original
data. It learns features from the neighborhood of the
investigated pixel to represent the whole local homogeneous
area of the image. In addition, to obtain more accurate
representation of the investigated pixel, a weighting scheme
is adopted based on the neighboring pixels, where the
weights are determined by the spectral similarity between
the neighboring pixels and the investigated pixel. The effectiveness
of our method is evaluated by the experiments
on two hyperspectral datasets and the results show that our
proposed method has better performance.
U2 - 10.1109/LGRS.2017.2737823
DO - 10.1109/LGRS.2017.2737823
M3 - Article
JO - IEEE Geoscience and Remote Sensing Letters
JF - IEEE Geoscience and Remote Sensing Letters
SN - 1545-598X
ER -