Transferred Multi-Perception Attention Networks for Remote Sensing Image Supe...

杨利霞

Transferred Multi-Perception Attention Networks for

Remote Sensing Image Super-Resolution

* `$ H( P# y4 KImage super-resolution (SR) reconstruction plays a key role in coping with the increasing
, B' z. k9 W, O" r) |demand on remote sensing imaging applications with high spatial resolution requirements. Though
+ a9 {$ E0 J% k% l3 [  ^9 Cmany SR methods have been proposed over the last few years, further research is needed to improve
- l4 ^" g2 |$ o8 Z1 p1 u1 vSR processes with regard to the complex spatial distribution of the remote sensing images and the
diverse spatial scales of ground objects. In this paper, a novel multi-perception attention network
( i( ^0 t4 @: E7 H$ F0 [6 S2 D(MPSR) is developed with performance exceeding those of many existing state-of-the-art models.
By incorporating the proposed enhanced residual block (ERB) and residual channel attention group
(RCAG), MPSR can super-resolve low-resolution remote sensing images via multi-perception learning
and multi-level information adaptive weighted fusion. Moreover, a pre-train and transfer learning
0 V) B2 v, n) mstrategy is introduced, which improved the SR performance and stabilized the training procedure.
Experimental comparisons are conducted using 13 state-of-the-art methods over a remote sensing
dataset and benchmark natural image sets. The proposed model proved its excellence in both objective
  _/ P3 S5 ]/ x$ H- w; L) X1 vcriterion and subjective perspective.


# C* x& `# ]9 d. ?  q, u7 O