A collection of information about Data Augmentation libraries for Image Processing.

CLoDSA

last updated on 10/10/2023

Repository Documentation Transformations Paper
Supported languages:
  • Python (2.7, 3.6)
Supported ML libraries:
  • Keras
Requirements:
  • absl-py==0.11.0
  • argon2-cffi==20.1.0
  • astunparse==1.6.3
  • async-generator==1.10
  • attrs==20.3.0
  • backcall==0.2.0
  • bleach==3.3.0
  • cached-property==1.5.2
  • cachetools==4.2.0
  • certifi==2020.12.5
  • cffi==1.14.4
  • chardet==4.0.0
  • clodsa==1.2.42
  • colorama==0.4.4
  • commentjson==0.9.0
  • cycler==0.10.0
  • decorator==4.4.2
  • defusedxml==0.6.0
  • entrypoints==0.3
  • flatbuffers==1.12
  • future==0.18.2
  • gast==0.3.3
  • google-auth==1.24.0
  • google-auth-oauthlib==0.4.2
  • google-pasta==0.2.0
  • grpcio==1.32.0
  • h5py==2.10.0
  • idna==2.10
  • imageio==2.9.0
  • imgaug==0.4.0
  • importlib-metadata==3.3.0
  • imutils==0.5.3
  • ipykernel==5.4.2
  • ipython==7.19.0
  • ipython-genutils==0.2.0
  • ipywidgets==7.5.1
  • jedi==0.17.2
  • Jinja2==2.11.3
  • joblib==1.0.0
  • jsonschema==3.2.0
  • jupyter==1.0.0
  • jupyter-client==6.1.7
  • jupyter-console==6.2.0
  • jupyter-core==4.7.0
  • jupyterlab-pygments==0.1.2
  • Keras==2.4.3
  • Keras-Preprocessing==1.1.2
  • kiwisolver==1.3.1
  • lark-parser==0.7.8
  • mahotas==1.4.11
  • Markdown==3.3.3
  • MarkupSafe==1.1.1
  • matplotlib==3.3.3
  • mistune==0.8.4
  • nbclient==0.5.1
  • nbconvert==6.0.7
  • nbformat==5.0.8
  • nest-asyncio==1.4.3
  • networkx==2.5
  • notebook==6.1.6
  • numpy==1.19.2
  • oauthlib==3.1.0
  • opencv-python==4.4.0.46
  • opt-einsum==3.3.0
  • packaging==20.8
  • pandas==1.1.5
  • pandocfilters==1.4.3
  • parso==0.7.1
  • pickleshare==0.7.5
  • Pillow==8.1.1
  • progressbar2==3.53.1
  • prometheus-client==0.9.0
  • prompt-toolkit==3.0.8
  • protobuf==3.14.0
  • pyasn1==0.4.8
  • pyasn1-modules==0.2.8
  • pycparser==2.20
  • Pygments==2.7.3
  • pyparsing==2.4.7
  • pyrsistent==0.17.3
  • python-dateutil==2.8.1
  • python-utils==2.4.0
  • pytz==2020.4
  • PyWavelets==1.1.1
  • pywin32==300
  • pywinpty==0.5.7
  • PyYAML==5.4
  • pyzmq==20.0.0
  • qtconsole==5.0.1
  • QtPy==1.9.0
  • requests==2.25.1
  • requests-oauthlib==1.3.0
  • rsa==4.6
  • scikit-image==0.18.1
  • scikit-learn==0.24.0
  • scipy==1.5.4
  • Send2Trash==1.5.0
  • Shapely==1.7.1
  • six==1.15.0
  • tensorboard==2.4.0
  • tensorboard-plugin-wit==1.7.0
  • tensorflow==2.4.0
  • tensorflow-estimator==2.4.0
  • termcolor==1.1.0
  • terminado==0.9.1
  • testpath==0.4.4
  • threadpoolctl==2.1.0
  • tifffile==2020.12.8
  • tornado==6.1
  • traitlets==5.0.5
  • typing-extensions==3.7.4.3
  • urllib3==1.26.3
  • wcwidth==0.2.5
  • webencodings==0.5.1
  • Werkzeug==1.0.1
  • widgetsnbextension==3.5.1
  • wrapt==1.12.1
  • zipp==3.4.0
Transformations Description Tags
1.

Smoothes the image using an average filter.

 BLUR
Original Augmented 
transformation = createTechnique("average_blurring", {"kernel": 5})
Original Augmented 
transformation = createTechnique("average_blurring", {"kernel": 11})
2.

Applies bilateral blurring to the image.

 BLUR
Original Augmented 
transformation = createTechnique("bilateral_blurring", {"diameter": 17,
                                 "sigmaColor": 31, "sigmaSpace": 11})
3.

Blurs an image using the normalized box filter.

 BLUR
Original Augmented 
transformation = createTechnique("blurring", {"ksize": 5})
Original Augmented 
transformation = createTechnique("blurring", {"ksize": 11})
4.

Crops pixels at the sides of the image.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("crop",{"percentage": 0.7,
                                 "startFrom": "TOPRIGHT"})
5.

Sets some pixels in the image to zero.

 DROPOUT
Original Augmented 
transformation = createTechnique("dropout",{"percentage": 0.05})
Original Augmented 
transformation = createTechnique("dropout",{"percentage": 0.5})
6.

Applies elastic deformation as explained in the paper P. Simard, D. Steinkraus, and J. C. Platt. Best practices for convolutional neural networks applied to visual document analysis. Proceedings of the 12th International Conference on Document Analysis and Recognition (ICDAR’03) vol. 2, pp. 958–964. IEEE Computer Society. 2003.

 DISTORTION
Original Augmented 
# alpha controlla l'intensità della deformazione
# sigma è il coefficiente di elasticità
transformation = createTechnique("elastic",{"alpha": 34,"sigma": 4})
Original Augmented
Original Augmented 
transformation = createTechnique("elastic",{"alpha": 34,"sigma": 3.5})
7.

Applies histogram equalization to the image.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("equalize_histogram",{})
8.

Flips the image horizontally, vertically or both.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("flip",{"flip": 0})
Original Augmented 
transformation = createTechnique("flip",{"flip": 1})
Original Augmented 
transformation = createTechnique("flip",{"flip": -1})
9.

Applies gamma correction to the image.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("gamma",{"gamma": 0.3})
Original Augmented 
transformation = createTechnique("gamma",{"gamma": 1.7})
10.

Blurs an image using a Gaussian filter.

 BLUR
Original Augmented 
transformation = createTechnique("gaussian_blur", {"kernel": 9})
11.

Adds Gaussian noise to the image.

 NOISE
Original Augmented 
transformation = createTechnique("gaussian_noise", {"mean": 0,"sigma": 30})
12.

Inverts all values in images, i.e. sets a pixel from value v to 255-v.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("invert",{})
13.

Blurs an image using the median filter.

 BLUR
Original Augmented 
transformation = createTechnique("median_blur", {"kernel": 9})
14.

Modifies the values in the blue channel based on the provided parameter.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_blue", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_blue", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_blue", {"power": 4})
15.

Modifies the values in the green channel based on the provided parameter.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_green", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_green", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_green", {"power": 4})
16.

Raises the hue value.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_hue", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_hue", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_hue", {"power": 4})
17.

Modifies the values in the red channel based on the provided parameter.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_red", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_red", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_red", {"power": 4})
18.

Modifies the saturation based on the provided parameter.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_saturation", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_saturation", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_saturation", {"power": 4})
19.

Modifies the value based on the provided parameter.

 PHOTOMETRY
Original Augmented 
transformation = createTechnique("raise_value", {"power": 0.26})
Original Augmented 
transformation = createTechnique("raise_value", {"power": 0.9})
Original Augmented 
transformation = createTechnique("raise_value", {"power": 4})
20.

Resizes the image.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("resize", {"percentage" : 0.3,
                                 "method":"INTER_NEAREST"})
21.

Rotates the image.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("rotate", {"angle" : 45})
22.

Adds salt and pepper noise to the image.

 NOISE
Original Augmented 
transformation = createTechnique("salt_and_pepper", {"low": 50,"up": 0})
Original Augmented 
transformation = createTechnique("salt_and_pepper", {"low": 0,"up": 50})
Original Augmented 
transformation = createTechnique("salt_and_pepper", {"low": 50,"up": 50})
23.

Sharpens the image.

 SHARPNESS
Original Augmented 
transformation = createTechnique("sharpen", {})
24.

Shears the image.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("shearing", {"a": 0.5})
Original Augmented 
transformation = createTechnique("shearing", {"a": -0.3})
25.

Shifts input image channels in the range given (value between 0 and 1).

 PHOTOMETRY
26.

Translates the image.

 BASIC GEOMETRY
Original Augmented 
transformation = createTechnique("translation", {"x": 40,"y": 10})
Original Augmented 
transformation = createTechnique("translation", {"x": -100,"y": -300})