Images to znyp format

It is mainly used to create images dataset for training, which supports RGB-image and one-channel grey image

example

image(JPG / PNG) files:

├── resize_image.py
├── load_znyp_format.py
├── convert_images_to_znyp_format.py
├── README.md
├── znyp_dataset
└── classes
    ├── 0
    │   ├── gesture-1.jpg
    │   └── gesture-2.jpg
    ├── 1
    │   ├── gesture-3.jpg
    │   └── gesture-4.jpg
    ├── 2
    │   └── gesture-5.jpg
   

prerequisites

depend on python3.5

• pillow(PIL)
• numpy
• matplotlib

usage

1. put your images in the class folder
2. modify the configuration that matches the images in convert_images_to_znyp_format.py
3. run the python script, which will transform you images into znyp format

  python convert_images_to_znyp_format.py

you can also load and visualize the dataset using python script

     python load_znyp_format.py