Received January 7, 2020, accepted January 15, 2020, date of publication January 20, 2020, date of current version

January 30, 2020.

Digital Object Identifier 10.1109/ACCESS.2020.2967804

Super-Resolution Reconstruction Method

of Remote Sensing Image Based on
Multi-Feature Fusion
ZHI-XING HUANG 1,3,4 AND CHANG-WEI JING2
1 Zhejiang Mariculture Research Institute, Zhejiang Key Lab of Exploitation and Preservation of Coastal Bio-resources, Wenzhou, 325000, China
2 Hangzhou Normal University, Hangzhou, 310000, China
3 Zhejiang College of Security Technology, Wenzhou, 325000, China
4 Zhejiang Yijia Geographic Information Technology Co., Ltd., Hangzhou, 310000, China

Corresponding author: Zhi-Xing Huang (635445367@qq.com)

This work was supported in part by the Key Special Projects through the Provincial Scientific Research Institutes Program of Zhejiang
Province, China, under Grant 2014F50022, and in part by the Construction of Agricultural Science Park Program of Zhejiang Province,
China, under Grant 2019E70002.

ABSTRACT The acquisition of remote sensing images is affected by imaging equipment and environmental
conditions. Usually on lower performance devices, the resolution of the acquired images is also low. Among
many methods, the super-resolution reconstruction method based on generative adversarial networks has
obvious advantages over previous network models in reconstructing image texture details. However, it is
found in experiments that not all of these reconstructed textures exist in the image itself. Aiming at the
problem of whether the texture details of the reconstructed image are accurate and clear, we propose a
super-resolution reconstruction method combining wavelet transform and generative adversarial network.
Using wavelet multi-resolution analysis, training wavelet decomposition coefficients in the generative adver-
sarial network can effectively improve the local detail information of the reconstructed image. Experimental
results show that our method can effectively reconstruct more natural image textures and make the images
more visually clear. In the remote sensing image test set, the four indicators of the algorithm, peak signal
to noise ratio (PSNR), structural similarity (SSIM), Feature Similarity (FSIM) and Universal Image Quality
(UIQ) are slightly better than the algorithms mentioned in the article.

INDEX TERMS Remote sensing image, super-resolution, self-correlation, image texture.

I. INTRODUCTION reconstruction technology is widely used in medical imaging,

With the continuous development of digital image process-
ing technology, people have higher and higher requirements
for image quality. In practical applications, high-resolution
images have more detailed information, which is more con-
ducive to processing and analysis. However, due to the limi-
tation of the quality of the imaging equipment, the generated
images usually cannot meet people’s needs. In order to obtain
high-resolution and clear images, it is usually to improve the
quality of hardware equipment, but this method is not only
complicated but also expensive. Therefore, super-resolution
reconstruction technology has become a research hotspot
in the field of image processing as an image processing
technology that processes low-resolution images by comput-
ers to reconstruct high-resolution images. Super-resolution
The associate editor coordinating the review of this manuscript and
approving it for publication was Wenming Cao . values of known neighboring pixels to generate the gray
satellite imagery, aerial remote sensing, criminal investiga-
tion and other fields. The use of super-resolution recon-
struction technology in medical images can help doctors
make more accurate medical diagnosis [1]–[4]; The use of
super-resolution reconstruction technology in satellite [5]
and aerial remote sensing [6] detection can obtain more
complex geographical environments Details; In the field of
criminal investigation super-resolution reconstruction tech-
nology can help provide more accurate information about sus-
pects and provide more effective means for the investigation
process [7].
The current super-resolution reconstruction methods can
be divided into three categories: interpolation-based methods,
reconstruction-based methods, and learning-based meth-
ods. The interpolation-based method is the earliest super-
resolution reconstruction method [8], which uses the gray

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0/
18764 VOLUME 8, 2020
Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion
values of the pixels to be interpolated. This method has
low complexity and fast speed, but the interpolation results
are easy. Causes artificial effects such as sawtooth, ringing
interference; The reconstruction-based method [9] is based
on a specific degradation model to provide constraints on
high-resolution image reconstruction based on the observed
low-resolution image sequence, and then fuses different
information of the same scene to obtain high quality. The
reconstruction results can better suppress the artificial effects,
but also cause the loss of detailed information, and the method
is complicated in operation, difficult to guarantee accuracy,
and low in efficiency; With the optimization of processor
performance, convenient conditions have been provided for
the field of big data and artificial intelligence, and deep learn-
ing applications have become more widespread [10], [11].
Learning-based algorithms are currently hotspots in the
field of super-resolution [12], the algorithm learns the map-
ping relationship between the high-resolution image and the
low-resolution image by extracting features, and finally real-
izes image reconstruction.
The super-resolution reconstruction method based on deep
learning is an emerging method, which has advantages over
traditional methods in terms of accuracy and clarity. However,
some super-resolution reconstruction algorithms based on
deep learning take minimizing mean square error as an opti-
mization goal [13]–[16]. Although this method can improve
the objective evaluation index of the reconstructed image,
it cannot effectively reconstruct the detailed texture of the
image. Ledig et al. proposed (Super-Resolution Using a Gen-
erative Adversarial Network, SRGAN) to apply a generative
adversarial network to the super-resolution reconstruction
field [17]. The reconstructed image texture is relatively clear,
but some details are not true compared to the real image.
In order to reconstruct more realistic image detail informa-
tion, we propose a super-resolution reconstruction algorithm
that combines wavelet transform technology and genera-
tive adversarial network technology from the perspective of
improving the high-frequency detail information of missing
images in low-resolution images. In this algorithm, we trans-
form the wavelet decomposition and reconstruction process
to a convolutional layer and a deconvolution layer in the
generated network structure. The decomposed wavelet coef-
ficients are trained in parallel independent subnets, and then
the output of the subnets is combined into a reconstructed
image by deconvolution. Finally, the discriminative network
is used to identify whether the output image of the generated
network is a high-resolution image. This method takes advan-
tage of the advantages of wavelet transform and generative
adversarial networks in the super-resolution field, so that the
reconstructed image has better texture information and is
more visually clear.
II. MODEL DESIGN
Our proposed super-resolution reconstruction algorithm
combines wavelet transform technology and generative
adversarial network. The advantage of generating adversarial
VOLUME 8, 2020
networks can reconstruct images with more realistic texture
details. In the network, the wavelet multi-resolution analysis
is used to train the wavelet decomposition coefficients to
improve the network’s understanding of the local details of
the image during the training process. Finally, it can recon-
struct the high frequency High-resolution images with better
detail and sharpness. This chapter will introduce the specific
operation of this method of image super-resolution recon-
struction in detail.
A. WAVELET TRANSFORM
In the method proposed in this paper, super-resolution recon-
struction of images is completed by training the image
wavelet decomposition coefficients during network training.
In the network, we use the changes between the channels of
the convolution layer to implement wavelet decomposition of
the feature image. The four coefficients decomposed are used
as the input of four independent subnets. Finally, the trained
wavelet coefficients are passed through a deconvolution layer
merge. The subnets are independent of each other and do not
affect each other. This method can realize multi-resolution
analysis of images in the generative adversarial network,
so that the network can both understand the entire image
and analyze the image details of the image. The network has
multiple aspects of understanding and deeper understanding
of the image, thereby effectively reconstructing the image
details more accurately and closer to the training results of
the original image.
When performing wavelet transform on an image, we con-
sider image I as a two-dimensional signal with index [a, b],
where I HR [a, b] is the pixel value in column b and a.
The two-dimensional signal I HR [a, b] can be considered as
a one-dimensional signal between a given b-th row I HR [a,:]
and a given a-th column I HR [:,b]. If the input low-resolution
image is I HR [a, b], we set the high-pass and low-pass filters
in the wavelet transform to h [k] and l [k], respectively.
In the two-dimensional discrete wavelet transform process,
we first perform high-pass and low-pass filtering and fre-
quency reduction processing on the b direction:
J −1
X
ILHR = I [a, 2b − j]l [j] (1)
J =0
J −1
X
IHHR = I [a, 2b − j]h [j] (2)
J =0
Then extend the a direction for high-pass and low-pass
filtering and frequency reduction processing.
J −1
X
HR
ILL = ILHR [2a − J , b]l [j] (3)
J =0
JX−1
HR
IHL = ILHR [2a − J , b]h [j] (4)
J =0
18765
 Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion

FIGURE 1. Generative adversarial model.

J −1
X parameters in the generation network and discriminant net-
HR
ILH = IHHR [2a − J , b]l [j] (5)
work. In this way, the generation network can continuously
J =0
optimize and get output results that are very similar to the
J −1
X original image. Therefore, super-resolution reconstruction
HR
IHH = IHHR [2a − J , b]h [j] (6)
based on image generation by adversarial networks can obtain
J =0
reconstructed images with good visual effects.
Among them, ILL HR obtained by equation (3) represents the

approximate coefficient of the original image, that is, the 1) GENERATING NETWORK
low-frequency component of the image; and IHL HR obtained by
HR HR obtained According to the definition of the generative adversarial net-
equation (4), ILH obtained by equation (5), and IHH
work, the hyperparameter δg of G is obtained by solving the
by equation (6) indicate The horizontal, vertical, and diagonal
smallest problem in Eq. (7).
detail coefficients of the image, that is, the high-frequency
components of the image.
In addition, when performing discrete wavelet decompo- 1 XN
sition and reconstruction of two-dimensional images, Haar θg = arg min lG [G (X )] (8)
N n=1
wavelet functions are often used as wavelet basis func-
tions. Haar wavelet function is the earliest tightly supported
In the formula, G consists of four functional modules:
orthogonal wavelet basis function. Its function is simple and
edge information extraction, residual dense learning and edge
widely used, especially it has good symmetry. Applying Haar
enhancement fusion, and image upsampling. The edge extrac-
wavelet basis function during training can effectively avoid
tion module obtains a preliminary edge image through a
image decomposition Generates phase distortion.
Laplacian sharpening filter [19], and transforms the prior
information of the edges through a 3 × 3 convolutional layer
B. GENERATIVE ADVERSARIAL NETWORK
into a highly discriminating Features; residual dense block
In generative adversarial networks, we need to define a gen-
(RDB) [20] can form a continuous memory mechanism, local
erative network and a discriminative network. The process of
feature fusion in RDB can adaptively learn from previous
generating adversarial network training is the process of con-
more effective features, which can stabilize the network train-
tinuously alternating optimization of the generation network
ing process and perceive loss In RDB, the Euclidean distance
and the discriminative network. The optimization process can
between image features is minimized by calculating the gap
be defined as:
in the feature space of the image, so that the generated image
min max EY ∼Ptrain(y) [log D(Y )] is less semantically different from the high-resolution remote
G D
sensing image feature and the generated image can be fitted to
+ EX ∼PG(x) [log(1−D(G(Y )))] (7)
high resolution Potential distribution of remote sensing image
Among them, G and D represent the generation network features. Definition i (•) represents the activation value
and discriminant network, and g and d represent the network of the output feature map after the ith convolutional layer of

18766 VOLUME 8, 2020

Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion
FIGURE 2. DOTA dataset test results.
RDB. The expression of the perceptual loss function is:
n
X (10)
lfeature = E(X ,Y )∼ptrain (X ,Y ) [ki (Y ) − i (G (X ))k] (9)
i=1
2) DISCRIMINATION NETWORK
D (Discriminant network) is used to estimate the probabil-
ity of false images from real sample data and G network,
where i is obtained by optimizing the minimum-maximum
problem in formula (7) in joint formula (8). D is mainly used
for binary classification to distinguish the authenticity of the
generated image. The network structure of D is shown in
Figure 1. The convolutional layer of s = 1 is used to extract
the shallow features of the image, and the convolutional layer
of s = 2 is used to reduce the image size. Then 4 dense con-
nection blocks (DDB) are used for dense connection learning
of the image features, and dense layers with PReLU and
Sigmoid activation functions are used to discriminate the true
VOLUME 8, 2020
and false attributes of image features [21].
ladv_G (X , Y ) = −EX ∼Ptrain(X ) log D (G (X ))
 
ladv_D (X , Y ) = EY ∼Ptrain(Y ) log D (Y )
 
+ EEX ∼ptrain (X ) log 1 − D (G (X )) (11)
 
Because the texture of the reconstructed image is too
unnatural after being enlarged, in order to deal with this
problem, we add a wavelet transform processing process in
the generation adversarial network, so that the network can
train the wavelet decomposition coefficients of the image,
thereby improving the high resolution The frequency detail
information makes the texture of the image more realistic.
Therefore, in this paper, our loss function adds the loss of
wavelet coefficients to the SRGAN perceptual loss function.
The loss function can be expressed as:
loss = l + lW
18767
 Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion

FIGURE 3. PSOD dataset test results.

The minimum mean square error lW of the wavelet decom- B. EVALUATION INDEX
position coefficient can be expressed as: The most commonly used indicators in the field of super-
 2  2  2 resolution reconstruction to evaluate the quality of the recon-
HR SR HR SR HR SR structed image are the peak signal-to-noise ratio and the
lW = a1 ILL − ILL +a2 ILH − ILH +a3 IHL − IHL
 2 Structural Similarity Index (SSIM) [22]. However, many
HR SR studies have shown that the mean square error and the
+ a4 IHH − IHH (12)
peak signal-to-noise ratio cannot accurately reflect the visual
quality of the image, so we also cited Feature Similarity
III. EXPERIMENT AND ANALYSIS (FSIM) [23], Universal Image Quality (UIQ) [25]–[27] to
A. DATA SET evaluate the reconstructed image quality. According to the
1) DOTA Human Visual System (HVS), which mainly understands
Contains 2806 remote sensing images, including: football images through their low-level features, Zhang et al. proposed
fields, helicopters, swimming pools, roundabouts, carts, FSIM. Phase consistency can extract highly image informa-
planes, ships, tanks, baseball fields, tennis courts, basketball tion features, but cannot respond to changes in image con-
courts, track and field fields, ports, bridges, and trolleys trast, so gradient features are introduced to extract contrast
15 categories. information. Phase consistency is used as the primary feature
of FSIM and gradient is used as the secondary feature. The
2) RSOD larger the value of FSIM, the higher the feature similarity
between the images, and the better the quality of the recon-
5364 targets including aircraft, playground, overpass, and
structed image. FSIM is represented as follows:
fuel tank.
SL (x) · PCm (x)
P
3) UCAS x∈8
FSIM = (13)
PCm (x)
P
A total of 7,000 images marked by the Pattern Recogni-
x∈8
tion and Intelligent System Development Laboratory of the
Chinese Academy of Sciences, including two types of targets: where 8 represents the pixel domain of the entire image,
cars and aircraft. and SPC (x) and SG (x) are expressed as the phase consistency

18768 VOLUME 8, 2020

Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion

FIGURE 4. UCAS dataset test results.

TABLE 1. On the three data sets of DOTA, RSOD, and UCAS, the average PSNR and SSIM values of the various super-resolution methods are enlarged 2, 3,
and 4 times.

TABLE 2. On the three data sets of DOTA, RSOD, and UCAS, the average FSIM and UIQ values of the various super-resolution methods are enlarged 2, 3,
and 4 times.

and gradient magnitude of the image, respectively. SL (x) and PCm (x) = max (PC1 (x), PC2 (x)) (15)
PCm (x) are expressed as: UIQ is an image quality evaluation indicator that combines
three factors, linear correlation loss, brightness distortion and
SL (x) = SPC (x)α · SG (x)β
   
(14) contrast distortion. The larger the value of UIQ, the higher the

VOLUME 8, 2020 18769

 Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion
correlation between the reconstructed image and the original
image in terms of linearity, brightness, and contrast, and the
better the quality of the reconstructed image. UIQ is defined
as follows:
σxy 2x̄ ȳ σx σy
UIQ = · · (16)
σx σy (x̄)2 + (ȳ)2 σx2 + σy2
Among them, the first term in formula (16) represents the
linear correlation coefficient between x and y. When the value
is 1, it means that x and y are linearly correlated. The second
term indicates the correlation of image brightness, and the
third term indicates the correlation of image contrast.
C. EXPERIMENTAL RESULTS AND ANALYSIS
In order to show the performance of the method in this paper,
we quantitatively compare the method in this paper with the
SRGAN method. Experiments are performed using DOTA,
RSOD, and UCAS data sets, and the obtained PSNR and
SSIM results of the reconstructed images are 2, 3, and 4 As
shown in Table 1. From the results in Table 1, it can be found
that the image quality assessment standards PSNR and SSIM
of the improved method proposed in this paper have a sig-
nificant improvement over most previous methods. In order
to evaluate the reconstructed image quality more accurately,
we used FSIM and UIQ to further analyze the experimental
results. As can be seen from Table 2, our proposed method
has achieved good results in both FSIM and UIQ evaluation
indicators.
IV. CONCLUSION
By studying super-resolution reconstruction algorithms based
on deep learning in recent years, we learned that generating
adversarial network models can reconstruct high-resolution
images with sharper textures. Because the wavelet transform
has the characteristics of decomposing the high-frequency
details and low-frequency in the image and expressing them
separately, in order to further improve the quality of the
reconstructed image, it is proposed to fuse the wavelet trans-
form into the structure of the adversarial network. Through
a series of experimental verifications, the super-resolution
reconstruction algorithm proposed in this paper can effec-
tively improve the shortcomings of the SRGAN algorithm’s
reconstructed image quality evaluation standard with lower
measured values, and PSNR, SSIM, FSIM, and UIQ all
increase. At the same time, the algorithm generates a recon-
structed image with realistic texture details. This method fully
reflects the advantages of this method in the field of super-
resolution reconstruction.
REFERENCES
[1] S. T. Alvarado, T. Fornazari, A. Cóstola, L. P. C. Morellato, and
T. S. F. Silva, ‘‘Drivers of fire occurrence in a mountainous Brazilian
cerrado savanna: Tracking long-term fire regimes using remote sensing,’’
Ecol. Indicators, vol. 78, pp. 270–281, Jul. 2017.
[2] J. S. Isaac and R. Kulkarni, ‘‘Super resolution techniques for medical
image processing,’’ in Proc. Int. Conf. Technol. Sustain. Develop. (ICTSD),
Feb. 2015, pp. 1–6.
18770
[3] H. Greenspan, ‘‘Super-resolution in medical imaging,’’ Comput. J., vol. 52,
no. 1, pp. 43–63, 2008.
[4] Y. Huang, L. Shao, and A. F. Frangi, ‘‘Simultaneous super-resolution and
cross-modality synthesis of 3D medical images using weakly-supervised
joint convolutional sparse coding,’’ in Proc. IEEE Conf. Comput. Vis.
Pattern Recognit. (CVPR), Jul. 2017, pp. 6070–6079.
[5] C. Wang, S. Jiang, H. Zhang, F. Wu, and B. Zhang, ‘‘Ship detection for
high-resolution SAR images based on feature analysis,’’ IEEE Geosci.
Remote Sens. Lett., vol. 11, no. 1, pp. 119–123, Jan. 2014.
[6] O. Cossairt, M. Gupta, and S. K. Nayar, ‘‘When does computational
imaging improve performance?’’ IEEE Trans. Image Process., vol. 22,
no. 2, pp. 447–458, Feb. 2013.
[7] J. Yang, Z. Lin, and S. Cohen, ‘‘Fast image super-resolution based on
in-place example regression,’’ in Proc. IEEE Conf. Comput. Vis. Pattern
Recognit., Jun. 2013, pp. 1059–1066.
[8] F. Zhou, W. Yang, and Q. Liao, ‘‘Interpolation-based image
super-resolution using multisurface fitting,’’ IEEE Trans. Image Process.,
vol. 21, no. 7, pp. 3312–3318, Jul. 2012.
[9] Z. Lin and H.-Y. Shum, ‘‘Fundamental limits of reconstruction-based
superresolution algorithms under local translation,’’ IEEE Trans. Pattern
Anal. Mach. Intell., vol. 26, no. 1, pp. 83–97, Jan. 2004.
[10] H. Zhang, Y. Fu, L.-B. Feng, Y. Zhang, and R. Hua, ‘‘Implementation of
hybrid alignment algorithm for protein database search on the SW26010
many-core processor,’’ IEEE Access, vol. 7, pp. 128054–128063, 2019.
[11] L.-Q. Zuo, H.-M. Sun, Q.-C. Mao, R. Qi, and R.-S. Jia, ‘‘Natural scene text
recognition based on encoder-decoder framework,’’ IEEE Access, vol. 7,
pp. 62616–62623, 2019.
[12] J. H. Zhou, C. Zhou, J. J. Zhu, and D. H. Fan, ‘‘Super-resolution reconstruc-
tion method of remote sensing image based on non-downsampling contour
wave transform,’’ J. Opt., vol. 35, no. 1, pp. 106–114, 2014.
[13] C. Dong, C. C. Loy, K. He, and X. Tang, ‘‘Learning a deep convolutional
network for image super-resolution,’’ in Proc. ECCV, 2014, pp. 184–199.
[14] J. Kim, J. K. Lee, and K. M. Lee, ‘‘Deeply-recursive convolutional network
for image super-resolution,’’ in Proc. IEEE Conf. Comput. Vis. Pattern
Recognit. (CVPR), Jun. 2016, pp. 1637–1645.
[15] C. Dong, C. C. Loy, K. He, and X. Tang, ‘‘Image super-resolution using
deep convolutional networks,’’ IEEE Trans. Pattern Anal. Mach. Intell.,
vol. 38, no. 2, pp. 295–307, Feb. 2016.
[16] W.-S. Lai, J.-B. Huang, N. Ahuja, and M.-H. Yang, ‘‘Deep Laplacian
pyramid networks for fast and accurate super-resolution,’’ in Proc. IEEE
Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 624–632.
[17] C. Ledig, L. Theis, F. Huszar, J. Caballero, A. Cunningham, A. Acosta,
A. Aitken, A. Tejani, J. Totz, Z. Wang, and W. Shi, ‘‘Photo-realistic
single image super-resolution using a generative adversarial network,’’
in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017,
pp. 105–114.
[18] X. Wang, ‘‘Laplacian operator-based edge detectors,’’ IEEE Trans. Pattern
Anal. Mach. Intell., vol. 29, no. 5, pp. 886–890, May 2007.
[19] K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, ‘‘Beyond a Gaussian
Denoiser: Residual learning of deep CNN for image denoising,’’ IEEE
Trans. Image Process., vol. 26, no. 7, pp. 3142–3155, Jul. 2017.
[20] Y. Zhang, Y. Tian, Y. Kong, B. Zhong, and Y. Fu, ‘‘Residual dense net-
work for image super-resolution,’’ in Proc. IEEE/CVF Conf. Comput. Vis.
Pattern Recognit., Jun. 2018, pp. 2472–2481.
[21] F. Li, H. Bai, L. Zhao, and Y. Zhao, ‘‘Dual-streams edge driven
encoder-decoder network for image super-resolution,’’ IEEE Access,
vol. 6, pp. 33421–33431, 2018.
[22] Y. Wang, L. Wang, H. Wang, and P. Li, ‘‘End-to-end image super-resolution
via deep and shallow convolutional networks,’’ IEEE Access, vol. 7,
pp. 31959–31970, 2019.
[23] Z. Lu, Z. Yu, P. Yali, L. Shigang, W. Xiaojun, L. Gang, and R. Yuan, ‘‘Fast
single image super-resolution via dilated residual networks,’’ IEEE Access,
vol. 7, pp. 109729–109738, 2019.
[24] A. K. Moorthy and A. C. Bovik, ‘‘Visual importance pooling for image
quality assessment,’’ IEEE J. Sel. Topics Signal Process., vol. 3, no. 2,
pp. 193–201, Apr. 2009.
[25] Z. Wang and A. Bovik, ‘‘A universal image quality index,’’ IEEE Signal
Process. Lett., vol. 9, no. 3, pp. 81–84, Mar. 2002.
[26] N. Damera-Venkata, T. Kite, W. Geisler, B. Evans, and A. Bovik, ‘‘Image
quality assessment based on a degradation model,’’ IEEE Trans. Image
Process., vol. 9, no. 4, pp. 636–650, Apr. 2000.
[27] H. Sheikh, M. Sabir, and A. Bovik, ‘‘A statistical evaluation of recent
full reference image quality assessment algorithms,’’ IEEE Trans. Image
Process., vol. 15, no. 11, pp. 3440–3451, Nov. 2006.
VOLUME 8, 2020
Z.-X. Huang, C.-W. Jing: Super-Resolution Reconstruction Method of Remote Sensing Image Based on Multi-Feature Fusion
ZHI-XING HUANG was born in Wenzhou,
Zhejiang, China, in 1984. He received the B.S.
degree from Chongqing University, in 2008, and
the master’s degree from the Beijing Institute of
Technology, in 2013. He is currently an Engineer
with the Zhejiang Marine Aquaculture Research
Institute, a Teacher with the Zhejiang Security
Vocational and Technical College, and a registered
Surveyor and Technical Consultant with Zhejiang
Yijia Geographic Information Technology Com-
pany, Ltd. His national research interests include research and application
of new technology of surveying and mapping remote sensing image, marine
environmental detection, and resource disaster prevention and control.
VOLUME 8, 2020
CHANG-WEI JING was born in Liaocheng,
Shandong, China, in 1983. He received the B.S.
and M.S. degrees from Jilin University and the
Ph.D. degree from Zhejiang University. Since
2013, he has been an Assistant Researcher with
the Zhejiang Provincial Key Laboratory of Urban
Wetland and Regional Change, Hangzhou, China.
He is the author of two books and more than
ten articles. His research interests include spatial
data processing and remote sensing for natural
resources and environment.
18771