Inception residual attention network for remote sensing image super-resolution

杨利霞

Inception residual attention network for remote sensingimage super-resolution

& l7 ~% K' l8 Z: Y& V, U& D0 _
, i, f% q6 e7 t3 \% X# b% P4 M, w
How to enhance the spatial resolution for a remote sensing image is
an important issue that we face. Many image super-resolution (SR)
techniques have been proposed for this purpose and deep con
volutional neural network (CNN) is the most effective approach in
; ^3 P# E  C1 x: y7 J! I( drecent years. However, we observe that most CNN-based SR meth
+ r& Z& y" V4 z' O$ R! ^ods treat low-frequency areas and high-frequency areas equally,
hence hindering the recovery of high-frequency information. In this
paper, we propose a network named inception residual attention
network (IRAN) to address this problem. Specifically, we propose
a spatial attention module to make the network adaptively learn
the importance of different spatial areas, so as to pay more atten
tion to the areas with high-frequency information. Furthermore, we
+ \/ o' _7 ]$ upresent an inception module to fuse local multilevel features, so as
2 I; t& B- e" Eto provide richer information for reconstructing detailed textures. In
& v2 P8 l8 c4 {* \5 ^order to evaluate the effectiveness of the proposed method, a large
number of experiments are performed on UCMerced-LandUse data
- A  h- M- h! h1 x8 Tset and the results show that the proposed method is superior to
the current state-of-the-art methods in both visual effects and
! _: h' O. T+ u4 q9 h! L4 Hobjective indicators.

( ]6 P  O7 ^3 P- \% ~
8 i2 s) o  D7 [& @3 f3 ]
2 M% i3 c6 _- z; k7 v
% ^/ |9 I0 Y7 o2 C$ @' O