Combining coordinate attention and generative adversarial network for image super-resolution reconstruction
An image super-resolution reconstruction model combining coordinate attention and gener-ative adversarial networks is proposed to address the problems of inadequate utilization of feature infor-mation,weak judgment of local details by VGG discriminators,and unstable training in the existing im-age super-resolution reconstruction model of generative adversarial networks.Firstly,a generator is constructed with residual blocks embedded with coordinate attention to aggregate features along both channel and spatial dimensions to extract features more adequately.The Dropout is also adjusted to join the network in such a way that it acts in the generator to improve the generalization ability of the model.Secondly,the discriminator is constructed with U-Net structure to output detailed pixel-by-pixel feed-back to obtain the local difference between the true and false images.Finally,spectral normalization regularization is introduced into the discriminator to stabilize the training of GAN.The experimental re-sults show that when the amplification factor is 4,the peak signal-to-noise ratio obtained on the bench-mark test sets Set5 and Set14 is increased by 1.75 dB on average,and the structural similarity is in-creased by 0.038 on average,which can reconstruct clearer and more realistic images with good visual effects.
万方数据
维普
