niftynet.io.misc_io module¶
-
niftynet.io.misc_io.
correct_image_if_necessary
(img)¶
-
niftynet.io.misc_io.
create_affine_pixdim
(affine, pixdim)¶ Given an existing affine transformation and the pixel dimension to apply, create a new affine matrix that satisfies the new pixel dimension :param affine: original affine matrix :param pixdim: pixel dimensions to apply :return:
-
niftynet.io.misc_io.
do_reorientation
(data_array, init_axcodes, final_axcodes)¶ Performs the reorientation (changing order of axes) :param data_array: Array to reorient :param ornt_init: Initial orientation :param ornt_fin: Target orientation :return data_reoriented: New data array in its reoriented form
-
niftynet.io.misc_io.
do_resampling
(data_array, pixdim_init, pixdim_fin, interp_order)¶ Performs the resampling :param data_array: Data array to resample :param pixdim_init: Initial pixel dimension :param pixdim_fin: Targeted pixel dimension :param interp_order: Interpolation order applied :return data_resampled: Array containing the resampled data
-
niftynet.io.misc_io.
expand_to_5d
(img_data)¶ Expands an array up to 5d if it is not the case yet :param img_data: :return:
-
niftynet.io.misc_io.
get_latest_subfolder
(parent_folder, create_new=False)¶
-
niftynet.io.misc_io.
image3
(name, tensor, max_outputs=3, collections=['summaries'], animation_axes=[1], image_axes=[2, 3], other_indices={})¶ Summary for higher dimensional images Parameters: name: string name for the summary tensor: tensor to summarize. Should be in the range 0..255.
By default, assumes tensor is NDHWC, and animates (through D) HxW slices of the 1st channel.collections: list of strings collections to add the summary to animation_axes=[1],image_axes=[2,3]
-
niftynet.io.misc_io.
image3_axial
(name, tensor, max_outputs=3, collections=['summaries'])¶
-
niftynet.io.misc_io.
image3_coronal
(name, tensor, max_outputs=3, collections=['summaries'])¶
-
niftynet.io.misc_io.
image3_sagittal
(name, tensor, max_outputs=3, collections=['summaries'])¶
-
niftynet.io.misc_io.
infer_ndims_from_file
(file_path)¶
-
niftynet.io.misc_io.
load_image
(filename)¶
-
niftynet.io.misc_io.
rectify_header_sform_qform
(img_nii)¶ Look at the sform and qform of the nifti object and correct it if any incompatibilities with pixel dimensions :param img_nii: :return:
-
niftynet.io.misc_io.
resolve_checkpoint
(checkpoint_name)¶
-
niftynet.io.misc_io.
save_data_array
(filefolder, filename, array_to_save, image_object=None, interp_order=3, reshape=True)¶ write image data array to hard drive using image_object properties such as affine, pixdim and axcodes.
-
niftynet.io.misc_io.
save_volume_5d
(img_data, filename, save_path, affine=array([[ 1., 0., 0., 0.], [ 0., 1., 0., 0.], [ 0., 0., 1., 0.], [ 0., 0., 0., 1.]]))¶ Save the img_data to nifti image :param img_data: 5d img to save :param filename: filename under which to save the img_data :param save_path: :param affine: an affine matrix. :return:
-
niftynet.io.misc_io.
set_logger
(file_name=None)¶
-
niftynet.io.misc_io.
split_filename
(file_name)¶
-
niftynet.io.misc_io.
squeeze_spatial_temporal_dim
(tf_tensor)¶ Given a tensorflow tensor, ndims==6 means: [batch, x, y, z, time, modality] this function removes x, y, z, and time dims if the length along the dims is one :return: squeezed tensor
-
niftynet.io.misc_io.
touch_folder
(model_dir)¶ This funciton returns the absolute path of model_dir if exists otherwise try to create the folder and returns the absolute path