Color-matcher batch processing

[SahPet]

I've just discovered color-matcher and find it potentially very useful for preprocessing histopathological datasets for deep learning. I can't, however, find a way to use it in batch mode - that is - is there any way to load more than one source image and/or more than one target image to process larger image datasets in batch?

[hahnec]

I am glad this tool may be useful for medical applications. So far, I have considered batch processing from the command line, which was supposed to work when providing the source directory as follows:

color-matcher -s './tests/data/' -r './tests/data/scotland_plain.png'

However, I figured that each output file gets overriden due to identical filename creation. I fixed this issue in v0.3.2, which is now available via pip install color-matcher==0.3.2

Similar to the Jupyter notebook example, you can alternatively use the API to iterate through your image stack via a single loop like the one below:

from color_matcher import ColorMatcher
from color_matcher.io_handler import load_img_file, save_img_file, FILE_EXTS
from color_matcher.normalizer import Normalizer
import os

img_ref = load_img_file('./tests/data/scotland_plain.png')

src_path = '.'
filenames = [os.path.join(src_path, f) for f in os.listdir(src_path)
                     if f.lower().endswith(FILE_EXTS)]

for i, fname in enumerate(filenames):
	img_src = load_img_file(fname)
	obj = ColorMatcher(src=img_src, ref=img_ref, method='mkl')
	img_res = obj.main()
	img_res = Normalizer(img_res).uint8_norm()
	save_img_file(img_res, os.path.join(os.path.dirname(fname), str(i)+'.png'))

Let me know whether this helps or if you still encounter issues.

Best,
Chris

[SahPet]

Thank you so much, Chris! I'll try it soon. Tried photoshops "match color" function, and I think color-matcher did a better job.

[Menante]

Thank you so much, It works perfectly! (I used it for own personal small software for my scansions of ancient books, which had to be normalized)

[hahnec]

Glad, it works on your side!