def __call__(self, filename):
"""
Args:
filename (str, list, tuple, file): path file or file-like object or a list of files.
"""
filename = ensure_tuple(filename)
img_array = list()
compatible_meta = dict()
for name in filename:
img = nib.load(name)
img = correct_nifti_header_if_necessary(img)
header = dict(img.header)
header["filename_or_obj"] = name
header["affine"] = img.affine
header["original_affine"] = img.affine.copy()
header["as_closest_canonical"] = self.as_closest_canonical
ndim = img.header["dim"][0]
spatial_rank = min(ndim, 3)
header["spatial_shape"] = img.header["dim"][1:spatial_rank + 1]
if self.as_closest_canonical:
img = nib.as_closest_canonical(img)
header["affine"] = img.affine
img_array.append(np.array(img.get_fdata(dtype=self.dtype)))
img.uncache()
if self.image_only:
continue
if not compatible_meta:
for meta_key in header:
meta_datum = header[meta_key]
# pytype: disable=attribute-error
if (type(meta_datum).__name__ == "ndarray"
and np_str_obj_array_pattern.search(
meta_datum.dtype.str) is not None):
continue
# pytype: enable=attribute-error
compatible_meta[meta_key] = meta_datum
else:
assert np.allclose(
header["affine"], compatible_meta["affine"]
), "affine data of all images should be same."
img_array = np.stack(img_array,
axis=0) if len(img_array) > 1 else img_array[0]
if self.image_only:
return img_array
return img_array, compatible_meta
def load_nifti(filename_or_obj,
as_closest_canonical=False,
image_only=True,
dtype=None):
"""
Loads a Nifti file from the given path or file-like object.
Args:
filename_or_obj (str or file): path to file or file-like object
as_closest_canonical (bool): if True, load the image as closest to canonical axis format
image_only (bool): if True return only the image volume, other return image volume and header dict
dtype (np.dtype, optional): if not None convert the loaded image to this data type
Returns:
The loaded image volume if `image_only` is True, or a tuple containing the volume and the Nifti
header in dict format otherwise
Note:
header['original_affine'] stores the original affine loaded from `filename_or_obj`.
header['affine'] stores the affine after the optional `as_closest_canonical` transform.
"""
img = nib.load(filename_or_obj)
img = correct_nifti_header_if_necessary(img)
header = dict(img.header)
header['filename_or_obj'] = filename_or_obj
header['original_affine'] = img.affine
header['affine'] = img.affine
header['as_closest_canonical'] = as_closest_canonical
if as_closest_canonical:
img = nib.as_closest_canonical(img)
header['affine'] = img.affine
if dtype is not None:
dat = img.get_fdata(dtype=dtype)
else:
dat = np.asanyarray(img.dataobj)
if image_only:
return dat
return dat, header
def read(self, data: Union[Sequence[str], str, Any]):
"""
Read image data from specified file or files, or set a Nibabel Image object.
Note that the returned object is Nibabel image object or list of Nibabel image objects.
`self._img` is always a list, even only has 1 image.
Args:
data: file name or a list of file names to read.
"""
self._img = list()
if isinstance(data, Nifti1Image):
self._img.append(data)
return data
filenames: Sequence[str] = ensure_tuple(data)
for name in filenames:
img = nib.load(name, **self.kwargs)
img = correct_nifti_header_if_necessary(img)
self._img.append(img)
return self._img if len(filenames) > 1 else self._img[0]
def read(self, data: Union[Sequence[str], str], **kwargs):
"""
Read image data from specified file or files.
Note that the returned object is Nibabel image object or list of Nibabel image objects.
Args:
data: file name or a list of file names to read.
kwargs: additional args for `nibabel.load` API, will override `self.kwargs` for existing keys.
More details about available args:
https://github.com/nipy/nibabel/blob/master/nibabel/loadsave.py
"""
img_: List[Nifti1Image] = list()
filenames: Sequence[str] = ensure_tuple(data)
kwargs_ = self.kwargs.copy()
kwargs_.update(kwargs)
for name in filenames:
img = nib.load(name, **kwargs_)
img = correct_nifti_header_if_necessary(img)
img_.append(img)
return img_ if len(filenames) > 1 else img_[0]
def read(self, data: Union[Sequence[str], str, Nifti1Image], **kwargs):
"""
Read image data from specified file or files, or set a Nibabel Image object.
Note that the returned object is Nibabel image object or list of Nibabel image objects.
`self._img` is always a list, even only has 1 image.
Args:
data: file name or a list of file names to read.
kwargs: additional args for `nibabel.load` API. more details about available args:
https://github.com/nipy/nibabel/blob/master/nibabel/loadsave.py
"""
self._img = list()
if isinstance(data, Nifti1Image):
self._img.append(data)
return data
filenames: Sequence[str] = ensure_tuple(data)
for name in filenames:
img = nib.load(name, **kwargs)
img = correct_nifti_header_if_necessary(img)
self._img.append(img)
return self._img if len(filenames) > 1 else self._img[0]
def test_affine(self):
test_img = nib.Nifti1Image(np.zeros((1, 2, 3)), np.eye(4) * 20.)
test_img = correct_nifti_header_if_necessary(test_img)
np.testing.assert_allclose(
test_img.affine, np.array([[20., 0., 0., 0.], [0., 20., 0., 0.], [0., 0., 20., 0.], [0., 0., 0., 20.]]))
def test_correct(self):
test_img = nib.Nifti1Image(np.zeros((1, 2, 3)), np.eye(4))
test_img.header.set_zooms((100, 100, 100))
test_img = correct_nifti_header_if_necessary(test_img)
np.testing.assert_allclose(
test_img.affine, np.array([[100., 0., 0., 0.], [0., 100., 0., 0.], [0., 0., 100., 0.], [0., 0., 0., 1.]]))
