A Super-Resolution Reconstruction Network of Space Target Images Based on Dual Regression and Deformable Convolutional Attention Mechanism

High-quality space target images are important for surveillance and attack defense confrontation. To obtain with higher resolution sharpness, this paper proposes an image super-resolution reconstruction network based on dual regression a deformable convolutional attention mechanism (DCAM). Firstly, the mapping is constrained by regression; secondly, convolution used to expand perceptual field extract high-frequency features of image; finally, calculate saliency channel domain spatial enhance useful suppress useless feature responses. The experimental results show that method outperforms comparison algorithm in both objective quality evaluation index localization accuracy dataset compared current mainstream algorithms.