14280746188

Committed 05 Apr 2025 09:50AM UTC coverage: 84.841% (-2.2%) from 87.002%

Build # 14280746188

Build Type

Pull #690

github

Committed by

web-flow

Commit Message

Merge d28da9558 into c7b17ed4a

Pull Request Pull Request #690: feature/add-qr-support

Run Details

318 of 410 new or added lines in 35 files covered. (77.56%)

5 existing lines in 2 files now uncovered.

2569 of 3028 relevant lines covered (84.84%)

3.24 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

96.97

/normcap/detection/ocr/recognize.py

"""Detect OCR tool & language and perform OCR on selected part of image."""

import logging
import tempfile
import time
from collections.abc import Iterable
from os import PathLike
from pathlib import Path
from typing import Optional, Union

import cv2
import numpy as np

from normcap.detection.ocr import enhance, tesseract, transformer
from normcap.detection.ocr.models import OEM, PSM, OcrResult, TessArgs

logger = logging.getLogger(__name__)


def _save_image_in_temp_folder(image: np.ndarray, postfix: str = "") -> None:
    """For debugging it can be useful to store the cropped image."""
    if logger.getEffectiveLevel() != logging.DEBUG:
        return

    temp_dir = Path(tempfile.gettempdir()) / "normcap"
    temp_dir.mkdir(exist_ok=True)

    now_str = time.strftime("%Y-%m-%d_%H-%M-%S", time.gmtime())
    file_name = f"{now_str}{postfix}.png"

    logger.debug("Save debug image as %s", temp_dir / file_name)
    cv2.imwrite(str(temp_dir / file_name), image)


def get_text_from_image(  # noqa: PLR0913
    tesseract_cmd: PathLike,
    languages: Union[str, Iterable[str]],
    image: np.ndarray,
    tessdata_path: Optional[PathLike] = None,
    parse: bool = True,
    resize_factor: Optional[float] = None,
    padding_size: Optional[int] = None,
) -> OcrResult:
    """Apply OCR on selected image section."""
    image = enhance.preprocess(image, resize_factor=resize_factor, padding=padding_size)
    _save_image_in_temp_folder(image, postfix="_enhanced")

    tess_args = TessArgs(
        tessdata_path=tessdata_path,
        lang=languages if isinstance(languages, str) else "+".join(languages),
        oem=OEM.DEFAULT,
        psm=PSM.AUTO,
    )
    logger.debug(
        "Run Tesseract on image of size %s with args:\n%s",
        image.shape[:2],
        tess_args,
    )
    ocr_result_data = tesseract.perform_ocr(
        cmd=tesseract_cmd, image=image, args=tess_args.as_list()
    )
    result = OcrResult(tess_args=tess_args, words=ocr_result_data, image=image)
    logger.debug("OCR detections:\n%s", ",\n".join(str(w) for w in result.words))

    if parse:
        result = transformer.apply(result)
        logger.debug("Parsed text:\n%s", result.parsed)

    return result

1	"""Detect OCR tool & language and perform OCR on selected part of image."""
2
3	import logging	4✔
4	import tempfile	4✔
5	import time	4✔
6	from collections.abc import Iterable	4✔
7	from os import PathLike	4✔
8	from pathlib import Path	4✔
9	from typing import Optional, Union	4✔
10
11	import cv2	4✔
12	import numpy as np	4✔
13
14	from normcap.detection.ocr import enhance, tesseract, transformer	4✔
15	from normcap.detection.ocr.models import OEM, PSM, OcrResult, TessArgs	4✔
16
17	logger = logging.getLogger(__name__)	4✔
18
19
20	def _save_image_in_temp_folder(image: np.ndarray, postfix: str = "") -> None:	4✔
21	"""For debugging it can be useful to store the cropped image."""
22	if logger.getEffectiveLevel() != logging.DEBUG:	4✔
NEW 23	return	×
24
25	temp_dir = Path(tempfile.gettempdir()) / "normcap"	4✔
26	temp_dir.mkdir(exist_ok=True)	4✔
27
28	now_str = time.strftime("%Y-%m-%d_%H-%M-%S", time.gmtime())	4✔
29	file_name = f"{now_str}{postfix}.png"	4✔
30
31	logger.debug("Save debug image as %s", temp_dir / file_name)	4✔
32	cv2.imwrite(str(temp_dir / file_name), image)	4✔
33
34
35	def get_text_from_image( # noqa: PLR0913	4✔
36	tesseract_cmd: PathLike,
37	languages: Union[str, Iterable[str]],
38	image: np.ndarray,
39	tessdata_path: Optional[PathLike] = None,
40	parse: bool = True,
41	resize_factor: Optional[float] = None,
42	padding_size: Optional[int] = None,
43	) -> OcrResult:
44	"""Apply OCR on selected image section."""
45	image = enhance.preprocess(image, resize_factor=resize_factor, padding=padding_size)	4✔
46	_save_image_in_temp_folder(image, postfix="_enhanced")	4✔
47
48	tess_args = TessArgs(	4✔
49	tessdata_path=tessdata_path,
50	lang=languages if isinstance(languages, str) else "+".join(languages),
51	oem=OEM.DEFAULT,
52	psm=PSM.AUTO,
53	)
54	logger.debug(	4✔
55	"Run Tesseract on image of size %s with args:\n%s",
56	image.shape[:2],
57	tess_args,
58	)
59	ocr_result_data = tesseract.perform_ocr(	4✔
60	cmd=tesseract_cmd, image=image, args=tess_args.as_list()
61	)
62	result = OcrResult(tess_args=tess_args, words=ocr_result_data, image=image)	4✔
63	logger.debug("OCR detections:\n%s", ",\n".join(str(w) for w in result.words))	4✔
64
65	if parse:	4✔
66	result = transformer.apply(result)	4✔
67	logger.debug("Parsed text:\n%s", result.parsed)	4✔
68
69	return result	4✔

dynobo / normcap / 14280746188

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous