IEEE Geoscience and Remote Sensing Magazine - December 2017 - 13

4
25
6
4,
09
6
4,
09
6

HYPERSPECTRAL IMAGE
.t.
ANALYSIS
ion
Forward/Inference
ict
ng
d
o
i
e
t
Hyperspectral sensors are characterPr
nta
ise
me
w
ized by hundreds of narrow spectral
g
l
e
Backward/Learning
Se
Pix
bands. This very high spectral resolution enables us to identify the materials
contained in the pixel via spectroscopic analysis. Analysis of hyperspectral
21
data is of great importance in many
practical applications, such as land
cover/use classification or change and
96
object detection. Because high-quality
hyperspectral satellite data are becom21
ing available (e.g., via the launch of
EnMAP, planned for 2020, and the
DESIS on the International Space Sta- FIGURE 3. The FCN architecture [7]. g.t.: ground truth.
tion, planned for 2018), hyperspectral
image analysis has been one of the most active research areas
layer, and an output layer-and directly classify hyperspecwithin the remote-sensing community over the last decade.
tral images in the spectral domain.
Inspired by the success of deep learning in computer viMakantasis et al. [26] exploited a two-dimensional
sion, preliminary studies have been carried out on deep learn(2-D) CNN to encode spectral and spatial information, foling in hyperspectral data analysis, which brings new momenlowed by a multilayer perceptron performing the actual
tum to this field. In the following, we review two application
classification. In [27], the authors attempted to carry out
cases, land cover/use classification and anomaly detection.
the classification of crop types using 1-D CNN and 2-D
CNN. They concluded that the 2-D CNNs can outperform
HYPERSPECTRAL IMAGE CLASSIFICATION
the 1-D CNNs, but some small objects in the final clasSupervised classification is probably the most active research
sification map provided by 2-D CNNs are smoothed and
area in hyperspectral data analysis. There is a vast literature
misclassified. To avoid overfitting, Zhao and Du [28] sugon this topic using conventional supervised machine-learngest a spectral-spatial-feature-based classification frameing models, such as decision trees, random forests, and supwork, which jointly makes use of a local-discriminant
port vector machines (SVMs) [20]. With the investigation of
embedding-based dimension-reduction algorithm and a
hyperspectral image classification [21], a major finding was
2-D CNN. In [21], the authors propose a self-improving
that various atmospheric scattering conditions, complicated
CNN model that combines a 2-D CNN with a fractionallight-scattering mechanisms, interclass similarity, and inorder Darwinian particle swarm optimization algorithm
traclass variability result in the hyperspectral imaging proto iteratively select the most informative bands suitable
cedure being inherently nonlinear. It is believed that, comfor training the designed CNN. Santara et al. [29] discuss
pared to the previously mentioned shallow models, deep
learning architectures are able to extract high-level, hierarchical, and abstract features, which are generally more robust
100+ (Predicted)
to the nonlinear processing.
The following sections discuss research on hyperspectral image classification.
78
71

SAE FOR HYPERSPECTRAL DATA CLASSIFICATION
A first attempt in this direction can be found in [22], where
the authors make use of an SAE to extract hierarchical features in the spectral domain. Subsequently, in [23], the authors employ DBN. Similarly, Tao et al. [24] use sparse SAEs
to learn an effective feature representation from input data;
then, the learned features are fed into a linear SVM for hyperspectral data classification.
SUPERVISED CNNs
In [25], the authors train a simple one-dimensional
(1-D) CNN that contains five layers-i.e., an input layer, a
convolutional layer, a max-pooling layer, a fully connected
DECEMBER 2017

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

(Sept. 2017)

23
3
2014

2015
2016
Publication Years

2017

FIGURE 4. The statistics for published papers related to deep learn-

ing in remote sensing [187].
13

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2017