Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

ImageAugmentation techniques for machine learning

This work allows an access to a set of image augmentation techniques. These techniques are not limited to simply flipping or rotation, but they include also more sophisticated approaches. These approach provide a little increment of accuracy on test set, with a relative substantial increase in computational effort.

To use them, call the method whenever necessary based on the image augmentation structure in use. For better performance, on-fly-augmentation is recommended.

Only RGB images are handled.

Methods

Sinusoidal

Deform the image following a sinusoidal function. In this specific case, a sin function.

Elastic_deformation

Apply a guassian based filter on the image and interpolate it.

Gaussian noise

Apply a gaussian noise on the image.

Crop_image

Crop the image.

To_gray

From an RGB to a scale of gray image

Salt_pepper_noise:

Add bright pixel on dark zones and vice versa.

Saturation

Add 20% saturation on the image.

Add_contrast

Add contrast to the image.

Blur

Add blur to the image.

Method1

Random black pixel setting on DCT image.

Method2

Disturbance of DCT image.

Method3

Random switch of elements on DCT image.

Pattern synthesis methods

It is provided a set of methods in order to create a new set of pattern. The process consists of 3 steps:

  1. Vectoritation of the patterns involved in the training set. Quantity is choosen by the developer.
  2. For each vector, find the more similar.
  3. Merge the vectors chosen at step 2.

The methods available for these steps are

  • color_vectoritation: transform RGB image to a vector based on its colour distribution.
  • texture_vectoritation: transform RGB image to a vector based on its LBP.
  • similarity: define a cosin similarity between two vectors.
  • euclidean_d: define euclidean distance between two vectors.
  • fusion: switch randomic color plane of two input RGB image.

Pattern synthesis has to be implemented using these methods directly on the data augmentation process, in order to allow more degrees of freedom.

Fusion of various methods provides an increment of ability to generalize of the net. AlexNet has been used for testing.

About

Image augmentation techniques for machine learning.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages