How to use the ocrmypdf.leptonica function in ocrmypdf

To help you get started, we’ve selected a few ocrmypdf examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github jbarlow83 / OCRmyPDF / tests / test_lept.py View on Github external
def test_colormap_backgroundnorm(resources):
    # Issue #262 - unclear how to reproduce exactly, so just ensure leptonica
    # can handle that case
    pix = lept.Pix.open(resources / 'baiona_colormapped.png')
    pix.background_norm()
github jbarlow83 / OCRmyPDF / tests / test_rotation.py View on Github external
ghostscript.rasterize_pdf(
            pdf,
            png,
            xres=100,
            yres=100,
            raster_device='pngmono',
            log=gslog,
            pageno=pageno,
            rotation=0,
        )

    rasterize(reference_pdf, reference_pageno, reference_png)
    rasterize(test_pdf, test_pageno, test_png)

    pix_ref = leptonica.Pix.open(reference_png)
    pix_test = leptonica.Pix.open(test_png)

    return leptonica.Pix.correlation_binary(pix_ref, pix_test)
github jbarlow83 / OCRmyPDF / tests / test_rotation.py View on Github external
png,
            xres=100,
            yres=100,
            raster_device='pngmono',
            log=gslog,
            pageno=pageno,
            rotation=0,
        )

    rasterize(reference_pdf, reference_pageno, reference_png)
    rasterize(test_pdf, test_pageno, test_png)

    pix_ref = leptonica.Pix.open(reference_png)
    pix_test = leptonica.Pix.open(test_png)

    return leptonica.Pix.correlation_binary(pix_ref, pix_test)
github jbarlow83 / OCRmyPDF / tests / test_rotation.py View on Github external
import img2pdf
import pikepdf
import pytest
from PIL import Image

from ocrmypdf import leptonica
from ocrmypdf.exec import ghostscript, tesseract
from ocrmypdf.pdfinfo import PdfInfo

# pytest.helpers is dynamic
# pylint: disable=no-member
# pylint: disable=w0612

pytestmark = pytest.mark.skipif(
    leptonica.get_leptonica_version() < 'leptonica-1.72',
    reason="Leptonica is too old, correlation doesn't work",
)

check_ocrmypdf = pytest.helpers.check_ocrmypdf
run_ocrmypdf = pytest.helpers.run_ocrmypdf


RENDERERS = ['hocr', 'sandwich']


def check_monochrome_correlation(
    outdir, reference_pdf, reference_pageno, test_pdf, test_pageno
):
    gslog = logging.getLogger()

    reference_png = outdir / f'{reference_pdf.name}.ref{reference_pageno:04d}.png'
github jbarlow83 / OCRmyPDF / tests / test_lept.py View on Github external
def test_crop(resources):
    pix = lept.Pix.open(resources / 'linn.png')
    foreground = pix.crop_to_foreground()
    assert foreground.width < pix.width