Data-Augmentation-Using-Keras-and-Python

Data augmentation is the process of increasing the number of training dataset. Keras library offers a simple solution for augmenting data with few lines of codes.

Practical Data Augmentation

the code in this repository performs augmentation by reading the dataset in three different folders and then save them in another folder. The process of data augmentation using Keras library is via ImageDataGenertor class.

Data Augmentation Outcome

This simple illustration of data augmentation using Keras library showcases the visual representation of data augmentation in practical.

Different Operations of Data Augmentation

A range of techniques are supported, as well as pixel scaling methods. Specifically, the five main types of data augmentation techniques for image data are;

1.) Image shifts via the width_shift_range and height_shift_range arguments.

2.) Image flips via the horizontal_flip and vertical_flip arguments.

3.) Image rotations via the rotation_range argument

4.) Image brightness via the brightness_range argument.

5. )Image zoom via the zoom_range argument.

Original Image

The origimnal inages are stored in three different folders

Augmented Images

The raw images are augmented and saved in a different folder.