VDL-SIM

The code is developed for real-time deap learning based SIM reconstruction with our VDL-SIM method, and is related to our paper:

"Video-level and high-fidelity super-resolution SIM reconstruction enabled by deep learning."
Advanced Imaging 1.1 (2024): 011001.

Contents

• Environment
• Requirement
• File structure
• Running guide

Environment

GPU: NVIDIA GeForce RTX 3050Ti
Tensorflow-gpu 2.10.0
Keras 2.10.0
CUDA 11.6
Python 3.9.5

Requirement

graphviz==0.20.1
h5py==3.7.0
imageio==2.22.4
keras==2.10.0
matplotlib==3.6.2
numpy==1.23.5
onnx==1.13.0
onnx-tf==1.10.0
opencv-python==4.6.0.66
pandas==1.5.3
Pillow==9.4.0
pyimagej==1.4.1
QtPy==2.3.0
scikit-image==0.19.3
scipy==1.10.1
tensorboard==2.10.1
tensorboardX==2.5.1
tensorflow-estimator==2.10.0
tensorflow-gpu==2.10.0
tf-slim==1.1.0
torch==1.13.1+cu116
torchaudio==0.13.1+cu116
torchvision==0.14.1+cu116
zipp==3.10.0

File structure

• ./models includes declaration of VDL-SIM model
• ./test includes some different SNR demo images of microtubules to test VDL-SIM model
• ./utils is the tool package of VDL-SIM
• ./weight place pre-trained VDL-SIM model here for testing
• ./models includes C++ interface for code

Running guide

• Download pre-trained models of VDL-SIM and place them in ./weight/
• Download test data and place them in ./test/images/. Also, you can prepare other testing data
• Open your terminal and run predict.py
• The output SR images will be saved in ./test/images/output_resu-SIM_weight-SIM/