Function equalize_hist

monai/transforms/utils.py:1840–1873

Utility to equalize input image based on the histogram. If `skimage` installed, will leverage `skimage.exposure.histogram`, otherwise, use `np.histogram` instead. Args: img: input image to equalize. mask: if provided, must be ndarray of bools or 0s and 1s, and same

(
    img: np.ndarray, mask: np.ndarray | None = None, num_bins: int = 256, min: int = 0, max: int = 255
)

def equalize_hist(
    img: np.ndarray, mask: np.ndarray | None = None, num_bins: int = 256, min: int = 0, max: int = 255
) -> np.ndarray:
    """
    Utility to equalize input image based on the histogram.
    If `skimage` installed, will leverage `skimage.exposure.histogram`, otherwise, use
    `np.histogram` instead.

    Args:
        img: input image to equalize.
        mask: if provided, must be ndarray of bools or 0s and 1s, and same shape as `image`.
            only points at which `mask==True` are used for the equalization.
        num_bins: number of the bins to use in histogram, default to `256`. for more details:
            https://numpy.org/doc/stable/reference/generated/numpy.histogram.html.
        min: the min value to normalize input image, default to `0`.
        max: the max value to normalize input image, default to `255`.

    """

    orig_shape = img.shape
    hist_img = img[np.array(mask, dtype=bool)] if mask is not None else img
    if has_skimage:
        hist, bins = exposure.histogram(hist_img.flatten(), num_bins)
    else:
        hist, bins = np.histogram(hist_img.flatten(), num_bins)
        bins = (bins[:-1] + bins[1:]) / 2

    cum = hist.cumsum()
    # normalize the cumulative result
    cum = rescale_array(arr=cum, minv=min, maxv=max)

    # apply linear interpolation
    img = np.interp(img.flatten(), bins, cum)
    return img.reshape(orig_shape)


class Fourier: