This repo contains code for BSRPCA, a method to do ULM processing by jointly perform deconvolution, tissue filtering and super resolution at the same time.
This repo contains code for **BSRPCA**, a method to do ULM processing by jointly perform deconvolution, tissue filtering and super resolution at the same time.
## Getting started
Please refer to the .m files for running code for experiment.
BSRPCA_convergent.m refers to the function implementation of our method while the invivo_exp.m contains a quick example of how to use our method with a sample vivo dataset.
`BSRPCA_convergent.m` refers to the function implementation of our method while the `invivo_exp.m` contains a quick example of how to use our method with a sample vivo dataset.
Our method is designed to be compatible with and based on the LOTUS Software - PALA : https://github.com/AChavignon/PALA
...
...
@@ -14,6 +14,16 @@ Please refer to the above software Repository to download required dependencies
Other methods and functions are from: https://github.com/phamduonghung/fast_BDRPCA and https://github.com/ning22/Fast-Single-Image-Superresolution/tree/master
## Datasets and Results
The in-vivo dataset contains **80** blocks, each with a spatial resolution of **160x128** and **500** frames per block (some irrelevant parts were excluded). The data was acquired at UMR 1253, iBrain, University of Tours, France.
The rendering results (Spatial Resolution: 1601 x 1281) for invivo dataset could be found in the files below:
-`rs_ulm_full.mat`: the rendering result of RS-ULM
-`our_method.mat`: the rendering result of BSRPCA
## Authors and acknowledgment
Xuan-Hieu Le, Jean-Pierre Remeniéras, Denis Kouame, and Duong-Hung Pham
