Multiprocessing library for OCR with WinRT

pip install winrtocr

Tested against Windows 10 / Python 3.10 / Anaconda

The results are not as good as the ones from Tesseract or EasyOCR, but it is way faster.

from winrtocr import WinRTocr

from time import perf_counter

from a_cv_imwrite_imread_plus import open_image_in_cv







winrto = WinRTocr(cpus=5, language="en-US")

picslinks = [

    r"https://github.com/hansalemaos/screenshots/raw/main/pandsnesteddicthtml.png",

    r"https://github.com/hansalemaos/screenshots/raw/main/cv2_putTrueTypeText_000000.png",

    r"https://github.com/hansalemaos/screenshots/raw/main/cv2_putTrueTypeText_000008.png",

    r"https://github.com/hansalemaos/screenshots/raw/main/cv2_putTrueTypeText_000017.png",

]



picsunique = [open_image_in_cv(x, channels_in_output=4) for x in picslinks]

pics = []

for _ in range(100):

    pics.extend(picsunique)



start = perf_counter()

dfa = winrto.get_ocr_df(pics)

print(f"Multi: {perf_counter()-start}")



dfa2 = winrto.get_ocr_df_and_pics(pics)