Metadata-Version: 2.1
Name: SegSRGAN
Version: 1.1.3
Summary: Segmentation and super resolution GAN network
Home-page: https://github.com/koopa31/SegSRGAN/tree/develop
Author: Clément Cazorla
Author-email: clement.cazorla@univ-reims.fr
License: UNKNOWN
Platform: UNKNOWN
Description-Content-Type: text/markdown
Requires-Dist: numpy (==1.16.2)
Requires-Dist: SimpleITK (==1.2.0)
Requires-Dist: scipy (==1.2.0)
Requires-Dist: tensorflow (==1.13.1)
Requires-Dist: keras (==2.2.4)
Requires-Dist: pandas (==0.23.0)
Requires-Dist: progressbar2 (==3.39.2)
Requires-Dist: requests (==2.18.4)
Requires-Dist: wget (==3.2)

# SegSRGAN

This algorithm is based on the [method](https://hal.archives-ouvertes.fr/hal-01895163) proposed by Chi-Hieu Pham in 2019.

## Installation

`pip install SegSRGAN`

## Perform a segmentation

`from SegSRGAN.SegSRGAN.Function_for_application_test_python3 import segmentation`

`segmentation(input_file_path, step, NewResolution, path_output_cortex, path_output_HR, weights_path, patch=None,
                 spline_order=3, by_batch=False, is_conditional=False)`

Where:
> * **input_file_path** is the path of the image to be super resolved and segmented 
> * **step** is the shifting step for the patches
> * **NewResolution** is the new z-resolution we want for the output image 
> * **path_output_cortex** output path of the segmented cortex
> * **path_output_HR** output path of the super resolution output image
> * **weights_path** is the path of the file which contains the pre-trained weights for the neural network
> * **patch** is the size of the patches
> * **spline_order** for the interpolation
> * **by_batch** is to enable the by-batch processing
> * **is_conditional** to perform a conditional GAN on the LR image resolution


## Segmentation of a set of images with several step and patch values

In order to facilitate the segmentation of several images, you can run SegSRGAN/SegSRGAN/job_model.py:

`python job_model.py --path
--patch --step --result_folder_name --weights_relative_path --is_conditional`

The list of the paths of the images to be processed must be stored in a csv file.

Where:

> * **path** Path of the csv file
> * **patch** list of patch sizes 
> * **step** list of steps 
> * **result_folder_name** Name of the folder containing the results
> * **is_conditional** Boolean to perform a conditional neural network with a condition on z-resolution

Example of syntax for step and patch setting:

--patch 64 128

--step 32 64,64 128

In this example we run steps 32 and 64 for patch 64 and steps 64 and 128 for patch 128. Be careful to respect the exact same spaces.



