Skew detection and correction in images containing text


Keywords
deskew, deskewing, scanning
License
MIT
Install
pip install deskew==1.5.1

Documentation

Deskew

Note: Skew is measured in degrees. Deskewing is a process whereby skew is removed by rotating an image by the same amount as its skew but in the opposite direction. This results in a horizontally and vertically aligned image where the text runs across the page rather than at an angle.

The return angle is between -45 and 45 degrees to don't arbitrary change the image orientation.

By using the library you can set the argument angle_pm_90 to True to have an angle between -90 and 90 degrees.

Skew detection and correction in images containing text

Image with skew Image after deskew
Image with skew Image after deskew

Installation

You can install deskew directly from pypi directly using the following comment

python3 -m pip install deskew

Or to upgrade to newer version

python3 -m pip install -U deskew

Cli usage

Get the skew angle:

deskew input.png

Deskew an image:

deskew --output output.png input.png

Lib usage

With scikit-image:

import numpy as np
from skimage import io
from skimage.color import rgb2gray
from skimage.transform import rotate

from deskew import determine_skew

image = io.imread('input.png')
grayscale = rgb2gray(image)
angle = determine_skew(grayscale)
rotated = rotate(image, angle, resize=True) * 255
io.imsave('output.png', rotated.astype(np.uint8))

With OpenCV:

import math
from typing import Tuple, Union

import cv2
import numpy as np

from deskew import determine_skew


def rotate(
        image: np.ndarray, angle: float, background: Union[int, Tuple[int, int, int]]
) -> np.ndarray:
    old_width, old_height = image.shape[:2]
    angle_radian = math.radians(angle)
    width = abs(np.sin(angle_radian) * old_height) + abs(np.cos(angle_radian) * old_width)
    height = abs(np.sin(angle_radian) * old_width) + abs(np.cos(angle_radian) * old_height)

    image_center = tuple(np.array(image.shape[1::-1]) / 2)
    rot_mat = cv2.getRotationMatrix2D(image_center, angle, 1.0)
    rot_mat[1, 2] += (width - old_width) / 2
    rot_mat[0, 2] += (height - old_height) / 2
    return cv2.warpAffine(image, rot_mat, (int(round(height)), int(round(width))), borderValue=background)

image = cv2.imread('input.png')
grayscale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
angle = determine_skew(grayscale)
rotated = rotate(image, angle, (0, 0, 0))
cv2.imwrite('output.png', rotated)

Debug images

If you get wrong skew angle you can generate debug images, that can help you to tune the skewing detection.

If you install deskew with pip install deskew[debug_images] you can get some debug images used for the skew detection with the function determine_skew_debug_images.

To start the investigation you should first increase the num_peaks (default 20) and use the determine_skew_debug_images function.

Then you can try to tune the following arguments num_peaks, angle_pm_90, min_angle, max_angle, min_deviation and eventually sigma.

Inspired by Alyn: https://github.com/kakul/Alyn

Contributing

Install the pre-commit hooks:

pip install pre-commit
pre-commit install --allow-missing-config