Multi-frame-image-super-resolution

Introduction of code file

*augmentation.py ---data augmentation *defshuffle.py ---shuffle images or image patches *config.py ---configuration hyperparameter

Run

• Installation ''' pip install tensorlayer==1.8.0 conda install tensorflow-gpu==1.8.0 pip install easydict ''' '''python config.VALID.img_path='your_image_folder/' config.TRAIN.img_path='your_image_folder/' config.VALID.logdir='your-tensorboard_folder/' '''
• Start Training ----- python main.py
• Start Testing ----- python main.py --mode=evaluate

Reference SRGAN_Wasserstein https://github.com/JustinhoCHN/SRGAN_Wasserstein

PS： if some parts of this project, please cite this paper "W. Li, Y. He, W. Kong, F. Gao, J. Wang, and G. Shi, "Enhancement of Retinal Image From Line-Scanning Ophthalmoscope Using Generative Adversarial Networks," IEEE Access 7, 99830–99841 (2019)."