Remote Sensing Image Colorization Based on Deep Neural Networks with Multi-Scale Residual Receptive Filed

To solve the problems of mistaken coloring and color bleeding in the current colorization methods,an end-to-end deep neural network is proposed to achieve remote sensing image colorization.First,the mul-ti-scale residual receptive filed net is introduced to extract the key features of source image.Second,a color information recovery network is con-structed by using U-Net,complex residual structure,attention mecha-nism,sequeeze-and-excitation and pixel-shuffle blocks to obtain color result.NWPU-RESISC45 dataset is chosen for model training and validation.Compared with other color methods,the PSNR value of the pro-posed method is increased by 6-10 dB on average and the SSIM value is increased by 0.05-0.11.In addition,the proposed method also achieves satisfactory color results on RSSCN7 and AID datasets.