3 d@sddlmZddlmZmZmZmZmZddlm Z GdddeZ GdddeZ Gd d d eZ dZ edde DZGdddeZGdddeZGdddeZddZdS))fname_presuffix)SimpleInterface TraitedSpecBaseInterfaceInputSpectraitsFile) LooseVersionc@sHeZdZeddddZeddddZejddZej ddddd d Z d S) RescaleInputSpecTzSkull-stripped image to rescale)exists mandatorydesczSkull-stripped reference imagez!Invert contrast of rescaled image)r ggI@zPercentile to use for reference to allow for outliers - 1 indicates the 1st and 99th percentiles in the input file will be mapped to the 99th and 1st percentiles in the reference; 0 indicates minima and maxima will be mapped)lowhighvalue usedefaultr N) __name__ __module__ __qualname__rin_fileref_filerZBoolinvertZRange percentilerr9/tmp/pip-build-7vycvbft/nipype/nipype/interfaces/image.pyr s r c@seZdZedddZdS)RescaleOutputSpecTzRescaled image)r r N)rrrrout_filerrrrrsrc@s eZdZdZeZeZddZdS)RescaleaXRescale an image Rescales the non-zero portion of ``in_file`` to match the bounds of the non-zero portion of ``ref_file``. Reference values in the input and reference images are defined by the ``percentile`` parameter, and the reference values in each image are identified and the remaining values are scaled accordingly. In the case of ``percentile == 0``, the reference values are the maxima and minima of each image. If the ``invert`` parameter is set, the input file is inverted prior to rescaling. Examples -------- To use a high-resolution T1w image as a registration target for a T2\* image, it may be useful to invert the T1w image and rescale to the T2\* range. Using the 1st and 99th percentiles may reduce the impact of outlier voxels. >>> from nipype.interfaces.image import Rescale >>> invert_t1w = Rescale(invert=True) >>> invert_t1w.inputs.in_file = 'structural.nii' >>> invert_t1w.inputs.ref_file = 'functional.nii' >>> invert_t1w.inputs.percentile = 1. >>> res = invert_t1w.run() # doctest: +SKIP cCsddl}ddl}|j|jj}|j}|j|jjj}|dk}|dk}|jjd|jjg} |j||| \} } |j||| \} } | | | | }|jjr| |n|| }|||| }|jjrdnd}t |jj||j d}|j ||j |j j|||jd<|S)NgY@Z_invZ _rescaled)suffixnewpathr)numpynibabelloadinputsrZ get_fdatarrrrcwd __class__affineheader to_filename_results)selfruntimenpnbimgdataZref_dataZin_maskZref_maskqZin_lowZin_highZref_lowZref_highZ scale_factorsignalZout_datarrrrr_run_interfaceBs& zRescale._run_interfaceN) rrr__doc__r input_specr output_specr3rrrrr srRLAPSIc cs~|]v}tD]l}tD]b}||ko2|ko2|knrdD]6}dD],}dD]"}dj||||||fVqLqDqas"rDc@s*eZdZeddddZejedddZdS)ReorientInputSpecTz Input image)r r r zTarget axis orientation)rr N) rrrrrrEnum _orientations orientationrrrrrElsrEc@s$eZdZedddZedddZdS)ReorientOutputSpecTzReoriented image)r r z1Affine transform from input orientation to outputN)rrrrr transformrrrrrIss rIc@s eZdZdZeZeZddZdS)ReorientaConform an image to a given orientation Flips and reorder the image data array so that the axes match the directions indicated in ``orientation``. The default ``RAS`` orientation corresponds to the first axis being ordered from left to right, the second axis from posterior to anterior, and the third axis from inferior to superior. For oblique images, the original orientation is considered to be the closest plumb orientation. No resampling is performed, and thus the output image is not de-obliqued or registered to any other image or template. The effective transform is calculated from the original affine matrix to the reoriented affine matrix. Examples -------- If an image is not reoriented, the original file is not modified .. testsetup:: >>> def print_affine(matrix): ... print(str(matrix).replace(']', ' ').replace('[', ' ')) >>> import numpy as np >>> from nipype.interfaces.image import Reorient >>> reorient = Reorient(orientation='LPS') >>> reorient.inputs.in_file = 'segmentation0.nii.gz' >>> res = reorient.run() >>> res.outputs.out_file 'segmentation0.nii.gz' >>> print_affine(np.loadtxt(res.outputs.transform)) 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. >>> reorient.inputs.orientation = 'RAS' >>> res = reorient.run() >>> res.outputs.out_file # doctest: +ELLIPSIS '.../segmentation0_ras.nii.gz' >>> print_affine(np.loadtxt(res.outputs.transform)) -1. 0. 0. 60. 0. -1. 0. 72. 0. 0. 1. 0. 0. 0. 0. 1. .. testcleanup:: >>> import os >>> os.unlink(res.outputs.out_file) >>> os.unlink(res.outputs.transform) cCsddl}ddl}ddlm}m}m}|jj}|j|}|j |j } ||jj } || | } || |j } t |jt dkr|j| } n t|| } | |k rd|jj j}t|||jd}| j|n|}t|d|jdd}|j|| d d ||jd <||jd <|S) Nr) axcodes2orntornt_transform inv_ornt_affz2.4.0_)rr z.matF)rr Zuse_extz%.08f)fmtrrJ)r!r"nibabel.orientationsrLrMrNr$rr#Zio_orientationr'rHshaper __version__Z as_reoriented_as_reoriented_backportlowerrr%r)Zsavetxtr*)r+r,r-r.rLrMrNfnameZorig_imgZ orig_orntZ targ_orntrJZ affine_xfm reorientedrZout_nameZmat_namerrrr3s.          zReorient._run_interfaceN) rrrr4rEr5rIr6r3rrrrrKzs:rKc sddl}ddl}ddlm}|jddgddgddggr>|S|j|j}|jj||j }|j |||j }t ||j rfdd|j jD}|j j||S)z1Backport of img.as_reoriented as of nibabel 2.4.0rN)rNrrcs(g|] }|dkrdnt|dfqS)Nr)int)r=Zorig_dim)orntrr sz+_as_reoriented_backport..)r!r"rQrNZ array_equalZapply_orientationZdataobjr'dotrRr&r( isinstanceZ Nifti1PairZ get_dim_infoZ set_dim_info) r/rYr-r.rNZt_arrZnew_affrWZnew_dimr)rYrrTs    rTN)r7r8r9)Zutils.filemaniprbaserrrrrr:r r rrr;tuplerGrErIrKrTrrrrs  ? g