Super-resolution Reconstruction Of Single Remote Sensing Image Based On Deep Learning

Super-resolution(SR),which restores a high-resolution(HR)image from single or sequential low-resolution(LR)images,is a widely applied technology in,especially in the remote sensing field.HR remote sensing images are increasingly sought with the rapid advancement of remote sensing technology in agriculture and forestry monitoring,urban planning,and military reconnaissance.However,traditional interpolation-based methods cannot achieve a satisfying effect,while reconstruction-based methods require pre-registration and are constrained by the lack of sequential images.In several modern learning-based methods,complicated network,considerable training time,and neglect of chrominance space still require improvement.To solve the problems above,a novel SR method combined with deep learning is proposed in this paper to achieve high-quality SR reconstruction of single remote sensing image,thereby overcoming traditional drawbacks,such as dependence on image sequences or registration.First,a network named PL-CNN that is based on a four-layer convolutional neural network(CNN)is designed,optimized with parametric rectified linear unit(PReLU)and local response normalization(LRN)layer after the first three convolutional layers.The PL-CNN is trained with an upscaling factor to obtain the SR model by taking the mean square error as the loss function.Then,for multi-band low-resolution SR images,a joint bilateral filtering under the guidance of the result is introduced to improve the edge details of the chrominance space after bicubic interpolation.A single-band image could be considered a special case of multiband image in which its reconstruction excludes the chrominance part.Results show that the proposed reconstruction of both multi-band and single-band remote-sensing images is superior to others at no-reference evaluation indexes with upscaling factors of 2,3,and 4.The mean clarity value improves by 2.5 standard units and the results at all-reference evaluation indexes are satisfying.The proposed method's results also display advantageous PSNR and efficiency,thereby achieving 2 dB better in PSNR than in bicubic interpolation algorithm and limiting the average training time to one-third or less than the other learning-based methods.The capability of joint bilateral filtering to remove the block effect and sharpen the edges is easily verified by observing the images of the chrominance space before and after filtering,therefore contributing to a more detailed and natural visual effect.After the post-processing of enhancement,the results of Spot-5 images can be close to the fused ones.Besides,the designed network has higher training efficiency and the ability of anti-overfitting.The proposed model can be optimized by increasing the number of training samples or selecting samples more relative.