def img_filt(fname):
# Load the thresholded CT
img = nib.load(fname)
# convert to nipy format for filtering
nipy_img = nipy.io.nifti_ref.nifti2nipy(img)
# build and apply gaussian filter
smoother = LinearFilter(nipy_img.coordmap, nipy_img.shape, 1)
smoothed_im = smoother.smooth(nipy_img)
# convert smoothed im back to nifti
nifti_img = nipy.io.nifti_ref.nipy2nifti(smoothed_im)
# get data array for saving
smoothed_CT = nifti_img.get_data()
smoothed_CT = np.array(smoothed_CT)
# save the smoothed thresholded ct
hdr = img.get_header()
affine = img.get_affine()
N = nib.Nifti1Image(smoothed_CT, affine, hdr)
new_fname = 'smoothed_' + fname
N.to_filename(new_fname)
def smooth_volume(nifti_file, smoothmm):
"""
@param nifti_file: string
@param smoothmm: int
@return:
"""
try:
img = load_image(nifti_file)
except Exception as exc:
raise Exception('Error reading file {0}.'.format(nifti_file), exc_info=True) from exc
if smoothmm <= 0:
return img
filter = LinearFilter(img.coordmap, img.shape)
return filter.smooth(img)
