This repository implements the analysis and application described in the paper
@article{Csoka2021,
author={\'Ad\'am Cs\'oka and Szilvia Eszter Simon and Tam\'as P\'eter Farkas and S\'andor Sz\'asz and Zolt\'an S\"ut\''o and \"Ors Petneh\'azy and Gy\"orgy Kov\'acs snd Imre Repa and Tam\'as Donk\'o},
title={ESTIMATION OF THE VALUABLE BROILER CHICKEN MEAT PARTS MASS FROM CT IMAGES USING ELASTIC REGISTRATION},
year={2024}
}
Preprint:
`000_extract_training_features.ipynb`
- segmentation by registration and feature extraction.`001_training.ipynb`
- feature subset and regressor parameter selection.`002_analysis.ipynb`
- the statistical analysis of the results, reproducing all the tables in the paper.`003_orchestration.ipynb`
- executes all notebooks
`config.py`
- high level configuration parameters.`requirements.txt`
- package requirements.`results.csv`
- raw results of the regression analysis with feature selection.`results.pickle`
- raw results of the regression analysis with feature selection in pickle format.`results.csv`
- raw results of the regression analysis without feature selection.`results.pickle`
- raw results of the regression analysis without feature selection in pickle format.`thigh.tex`
- collected results of the thigh in tex (typesetting language) format.`brest.tex`
- collected results of the brest in tex (typesetting language) format.`chicken_dissected_data.xlsx`
- results of the dissection study.
Clone the `maweight`
Python package (https://github.com/cseka7/maweight):
> git clone https://github.com/cseka7/maweight.git
Navigate into the root directory of the `maweight`
repository and issue
> pip install .
Clone this package (chicken_ct_weights):
> git clone https://github.com/cseka7/chicken_ct_weights.git
Navigate into the root directory of this package, and issue
> pip install -r requirements.txt
Download the CT images corresponding to the dissection study and the manual annotations from the link https://drive.google.com/file/d/1yz67G03E_avm-FGZ96m5gBxpHTy4DVRs/view?usp=sharing and extract its contents to the `data`
directory.
Update the paths in the file `config.py`
to match the environment the code is running in.
Start a jupyter server in the active environment by issuing
> jupyter notebook
And run the notebook `003_orchestration.ipynb`
to carry out all steps of the analysis.
Note that due to the large number of CT images and registered masks, the execution requires about 40Gb free space on the disk.