-
Download code and data of Holstein cattle from "https://github.com/ameybhole77/IIHC"
-
Upload data to matlab workspace. (matlab online in my case)
-
Run "BatchPreprocessing.m" program in matlab
- replace ';' with ':' in the code
- Sample folders will be created with 4 images - 2 segMasks and 2 segRGB images
-
Download this segmented data from matlab workspace into folder "Images_segmented".
-
Run "mypaint.py" program -
- This will read images from folder "Images_segmented".
- Then it will do inpainting and then resizing to (256*256)
- Next it will create a new folder "Images_inpaint_resized" contaning all output images.
-
Run "my_image_to_numpy.py" program -
- This will read images from folder "Images_inpaint_resized".
- Then it will split data into training and testing
- Then it will create a new folder "Images_numpy" storing data in numpy format.
- It will output shape of the numpy arrays. See screenshot - "ss_shape-of-training-testing.png"
nnboru / iihc Goto Github PK
View Code? Open in Web Editor NEWThis project forked from ameybhole/iihc