Study of synthesis images noise detection using 26 attributes

Jérôme BUISINE d4db4f7ca7 use of k fold and cv		4 lat temu
analysis	cb6026f2c7 Add of 26 features metric	5 lat temu
generate	c409191b1c update prediction script for new dataset structure	4 lat temu
macop @ b6eb599a16	f38dd28820 update gpu version params	4 lat temu
modules @ cebf2adbf1	0a05939b74 Update of optimization process using backups	5 lat temu
prediction	c409191b1c update prediction script for new dataset structure	4 lat temu
.gitignore	0a05939b74 Update of optimization process using backups	5 lat temu
.gitmodules	f38dd28820 update gpu version params	4 lat temu
LICENSE	dc0463b6b5 Project initialization	5 lat temu
README.md	30d257e0f7 update of the whole project to enable use of new dataset	4 lat temu
custom_config.py	e567b11e90 add svm gpu usage	4 lat temu
data_attributes.py	a4119a186e update kolmogorov attributes	4 lat temu
find_best_attributes.py	e88224dc08 update backup folder path	4 lat temu
find_best_filters.py	e88224dc08 update backup folder path	4 lat temu
models.py	d4db4f7ca7 use of k fold and cv	4 lat temu
requirements.txt	c52c6fae6c now use of macop Python package for optimization process	4 lat temu
train_model.py	c409191b1c update prediction script for new dataset structure	4 lat temu
train_model_attributes.py	b73b27ab44 add balanced data into SVC	4 lat temu
train_model_filters.py	b73b27ab44 add balanced data into SVC	4 lat temu

Noise detection using 26 attributes

Description

Noise detection on synthesis images with 26 attributes obtained using few filters.

Filters list:

pip install -r requirements.txt

You need database which respects this structure:

modules/*: contains all modules usefull for the whole project (such as configuration variables)
analysis/*: contains all jupyter notebook used for analysis during thesis
generate/*: contains python scripts for generate data from scenes (described later)
data_processing/*: all python scripts for generate custom dataset for models
prediction/*: all python scripts for predict new threshold from computed models
data_attributes.py: files which contains all extracted features implementation from an image.
custom_config.py: override the main configuration project of modules/config/global_config.py
train_model.py: script which is used to run specific model available.

data/*: folder which will contain all generated .train & .test files in order to train model.
data/saved_models/*: all scikit learn or keras models saved.
data/models_info/*: all markdown files generated to get quick information about model performance and prediction obtained after running run/runAll_*.sh script.
data/results/: This folder contains model_comparisons.csv file used for store models performance.