lyman.signals.smooth_surface¶
-
lyman.signals.smooth_surface(data_img, vert_img, fwhm, subject, surf='graymid', noise_img=None, inplace=False, subjects_dir=None)[source]¶ Smooth cortical voxels with Gaussian weighted surface distances.
- Parameters
- data_imgnibabel image
3D or 4D input data.
- vert_img4D nibabel image
Image where voxels have their corresponding surfave vertex ID or are -1 if they do not correspond to cortex. The first frame should have left hemisphere vertices and the second right hemisphere vertices.
- fwhmpositive float or None
Size of isotropic smoothing kernel, in mm, or None to return input (as float and possibly copied).
- subjectstring
Subject ID to locate data in data directory.
- surfstring
Name of the surface defining the mesh geometry.
- noise_imgnibabel image
Binary image defining voxels that should be interpolated out.
- inplace :bool
If True, overwrite data in data_img. Otherwise perform a copy.
- subjects_dirstring
Path to the Freesurfer data directory root; if absent, get from the SUBJECTS_DIR environment variable.
- Returns
- smooth_imgnibabel image
Image like
data_imgbut after smoothing.