Aiming to improve existing deep-learning based method to perform Super-Resolution Microscopy, model uses a tiling strategy, which takes advantage of parallelism provided by a GPU to speed up the network training process and impliments simple changes to the architecture of the generator and the discriminator of SRGAN. The model has been tested for cross domain dataset and has achieved significant results showing it's flexibility. For upscaling, model uses Nearest neighbor approach and for the discriminator, a Global Average Pooling (GAP) layer is added to the output of the last convolutional block.
Input variables : Microscopic Image
Output Variables : Image with enhanced resolution
Visit Model : github.com
Additional links : arxiv.org
| Model Category | : | Public |
| Date Published | : | October, 2020 |
| Healthcare Domain | : | Medical Technology |
| Code | : | github.com |
| Medical Imaging |
| Image Synthesis |