__all__ = ['smooth_image']
import math
from ants.decorators import image_method
from ants.internal import get_lib_fn
import ants
@image_method
def smooth_image(image, sigma, sigma_in_physical_coordinates=True, FWHM=False, max_kernel_width=32):
"""
Smooth an image
ANTsR function: `smoothImage`
Arguments
---------
image
Image to smooth
sigma
Smoothing factor. Can be scalar, in which case the same sigma is applied to each dimension, or a vector of length dim(inimage) to specify a unique smoothness for each dimension.
sigma_in_physical_coordinates : boolean
If true, the smoothing factor is in millimeters; if false, it is in pixels.
FWHM : boolean
If true, sigma is interpreted as the full-width-half-max (FWHM) of the filter, not the sigma of a Gaussian kernel.
max_kernel_width : scalar
Maximum kernel width
Returns
-------
ANTsImage
Example
-------
>>> import ants
>>> image = ants.image_read( ants.get_ants_data('r16'))
>>> simage = ants.smooth_image(image, (1.2,1.5))
"""
if image.components == 1:
outimage = image.clone()
if not isinstance(sigma, (tuple,list)):
sigma = [sigma]
if isinstance(sigma, (tuple, list)) and ((len(sigma) != image.dimension) and (len(sigma) != 1)):
raise ValueError('Length of sigma must be either 1 or the dimensionality of input image')
image_float = image.clone('float')
if FWHM:
sigma = [s/2.355 for s in sigma]
max_kernel_width = int(math.ceil(max_kernel_width))
smooth_image_fn = get_lib_fn('SmoothImage')
outimage = smooth_image_fn(image_float.pointer, sigma, sigma_in_physical_coordinates, max_kernel_width)
ants_outimage = ants.from_pointer(outimage)
return ants_outimage
else:
imagelist = ants.split_channels(image)
newimages = []
for image in imagelist:
newimage = smooth_image(image, sigma, sigma_in_physical_coordinates, FWHM, max_kernel_width)
newimages.append(newimage)
return ants.merge_channels(newimages)