cv3.processing.threshold

cv3.processing.threshold(img: ndarray, thr=127, max=None, type=None, rel=None)[source]

Apply threshold to a grayscale image.

Parameters:

  • img (numpy.ndarray) – Input grayscale image.

  • thr (int or float, optional) – Threshold value. Defaults to 127.

  • max (int or float, optional) – Maximum value to use with the thresholding. If None, defaults to 255. Defaults to None.

  • type (int or str, optional) – Threshold type. Can be an OpenCV threshold flag or string. Available string options: ‘binary’, ‘binary_inv’, ‘trunc’, ‘tozero’, ‘tozero_inv’. If None, defaults to opt.THRESHOLD_TYPE. Defaults to None.

  • rel (bool, optional) – Whether to use relative threshold value. Defaults to None.

Returns:

Thresholded image.

Return type:

numpy.ndarray

Raises:

AssertionError – If the input image is not a grayscale image.

Note

This function applies thresholding using the specified threshold type. Pixels are processed according to the chosen thresholding method.

Relative threshold values are in the range [0, 1] where 0 is the minimum and 1 is the maximum pixel value in the image.

Example

>>> import cv3
>>> import numpy as np
>>> # Create a simple grayscale image
>>> img = np.zeros((100, 100), dtype=np.uint8)
>>> img[25:75, 25:75] = 128  # Gray square
>>> # Apply threshold with default type
>>> thresh = cv3.threshold(img, 100)
>>> # Apply threshold with custom type
>>> thresh = cv3.threshold(img, 100, type='binary_inv')
>>> # Apply threshold with relative threshold value
>>> thresh = cv3.threshold(img, 0.5, rel=True)