Syntesis images noise detection using CNN approach
jbuisine 3d58035040 Merge branch 'release/v0.0.6' | преди 6 години | |
---|---|---|
img_train | преди 6 години | |
img_validation | преди 6 години | |
modules | преди 6 години | |
.gitignore | преди 6 години | |
README.md | преди 6 години | |
RESULTS.md | преди 6 години | |
classification_cnn_keras.py | преди 6 години | |
classification_cnn_keras_cross_validation.py | преди 6 години | |
classification_cnn_keras_svd.py | преди 6 години | |
config.json | преди 6 години | |
generate_dataset.py | преди 6 години | |
requirements.txt | преди 6 години |
pip install -r requirements.txt
Generate dataset (run only once time or clean data folder before) :
python generate_dataset.py
It will split scenes and generate all data you need for your neural network. You can specify the number of sub images you want in the script by modifying _NUMBER_SUBIMAGES variables.
There are 3 kinds of Neural Networks :
Note that the image input size need to change in you used specific size for your croped images.
After your built your neural network in classification_cnn_keras.py, you just have to run it :
classification_cnn_keras_svd.py --directory xxxx --output xxxxx --batch_size xx --epochs xx --img xx (or --image_width xx --img_height xx)
A config file in json is available and keeps in memory all image sizes available.
This project contains modules :
All these modules will be enhanced during development of the project
This git project uses git-flow implementation. You are free to contribute to it.