Improved Super-Resolution Convolution Neural Network for Large Images
This addresses a practical issue for users applying super-resolution to large images, though it appears incremental as it builds on existing SRCNN methods.
The paper tackled the problem of visible cutting lines in merged super-resolution images when using convolutional neural networks on large images, and proposed a refined SRCNN architecture with symmetric padding, random learning, and residual learning, achieving state-of-the-art performance in experiments.
Single image super-resolution (SISR) is a very popular topic nowadays, which has both research value and practical value. In daily life, we crop a large image into sub-images to do super-resolution and then merge them together. Although convolution neural network performs very well in the research field, if we use it to do super-resolution, we can easily observe cutting lines from merged pictures. To address these problems, in this paper, we propose a refined architecture of SRCNN with 'Symmetric padding', 'Random learning' and 'Residual learning'. Moreover, we have done a lot of experiments to prove our model performs best among a lot of the state-of-art methods.