Functions
ptr< MRImage >	npl::createMRImage (size_t ndim, const size_t *size, PixelT type)
	Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable. More...

ptr< MRImage >	npl::createMRImage (const std::vector< size_t > &dim, PixelT type)
	Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable. More...

ptr< MRImage >	npl::createMRImage (size_t ndim, const size_t size, PixelT type, void ptr, std::function< void(void *)> deleter)
	Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable. More...

ptr< MRImage >	npl::createMRImage (const std::vector< size_t > &dim, PixelT type, void ptr, std::function< void(void )> deleter)
	Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable. More...

std::ostream &	npl::operator<< (std::ostream &out, const MRImage &img)
	Writes out information about an MRImage. More...

void	npl::gaussianSmooth1D (ptr< MRImage > inout, size_t dim, double stddev)
	Gaussian smooths an image in 1 direction. Questionable whether it works. Seems to shift image. More...

ptr< MRImage >	npl::smoothDownsample (ptr< const MRImage > in, double sigma, double spacing=-1)
	Performs smoothing in each dimension, then downsamples so that pixel spacing is roughly equal to FWHM. More...

ptr< MRImage >	npl::resample (ptr< const MRImage > in, double spacing, double(window)(double, double)=hannWindow)
	Performs fourier resampling using fourier transform and the provided window function. More...

ptr< MRImage >	npl::shiftImage (ptr< MRImage > in, size_t len, double *vect)
	Uses fourier shift theorem to shift an image. More...

void	npl::writeComplex (std::string basename, ptr< const MRImage > in, bool absPhase=false)
	Writes a pair of images, one real, one imaginary or if absPhase is set to true then an absolute image and a phase image. More...

ptr< MRImage >	npl::fft_forward (ptr< const MRImage > in, const std::vector< size_t > &in_osize)
	Performs forward FFT transform in N dimensions. More...

ptr< MRImage >	npl::fft_backward (ptr< const MRImage > in, const std::vector< size_t > &in_osize)
	Performs inverse FFT transform in N dimensions. More...

int	npl::rotateImageShearKern (ptr< MRImage > inout, double rx, double ry, double rz, double(*kern)(double, double)=npl::lanczosKern)
	Rotates an image around the center using shear decomposition followed by kernel-based shearing. Rotation order is Rz, Ry, Rx, and about the center of the image. This means that 1D interpolation will be used. More...

int	npl::rotateImageShearFFT (ptr< MRImage > inout, double rx, double ry, double rz, double(*window)(double, double)=npl::sincWindow)
	Rotates an image around the center using shear decomposition followed by FFT-based shearing. Rotation order is Rz, Ry, Rx, and about the center of the image. More...

ptr< MRImage >	npl::rigidTransform (ptr< MRImage > in, double rx, double ry, double rz, double sx, double sy, double sz)
	Rigid Transforms an image. More...

ptr< MRImage >	npl::diffOfGauss (ptr< const MRImage > in, double sd1, double sd2)
	Computes difference of gaussians. More...

double	npl::overlapRatio (ptr< const MRImage > a, ptr< const MRImage > b)
	Computes the overlap of the two images' in 3-space. More...

ptr< MRImage >	npl::resampleNN (ptr< const MRImage > in, ptr< const MRImage > atlas, PixelT type=UNKNOWN_TYPE)
	Performs nearest neighbor resasmpling of input to atlas. More...

ptr< MRImage >	npl::resampleNN (ptr< const MRImage > in, double *newspace, PixelT type=UNKNOWN_TYPE)
	Performs nearest neighbor resasmpling of input to atlas. More...

ptr< MRImage >	npl::randImage (PixelT type, double mean, double sd, size_t x, size_t y, size_t z, size_t t)
	Create random image, with gaussian distribution. More...

Detailed Description

Function Documentation

ptr<MRImage> npl::createMRImage	(	size_t	ndim,
		const size_t *	size,
		PixelT	type
	)

Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable.

Parameters

ndim	number of image dimensions
size	size of image, in each dimension
type	Pixel type npl::PixelT

Returns: New image, default orientation

ptr<MRImage> npl::createMRImage	(	const std::vector< size_t > &	dim,
		PixelT	type
	)

Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable.

Parameters

dim	size of image, in each dimension, number of dimensions decied by length of size vector
type	Pixel type npl::PixelT

Returns: New image, default orientation

ptr<MRImage> npl::createMRImage	(	size_t	ndim,
		const size_t *	size,
		PixelT	type,
		void *	ptr,
		std::function< void(void *)>	deleter
	)

Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable.

Parameters

ndim	number of image dimensions
size	size of image, in each dimension
type	Pixel type npl::PixelT
ptr	Pointer to data block
deleter	function to delete data block

Returns: New image, default orientation

ptr<MRImage> npl::createMRImage	(	const std::vector< size_t > &	dim,
		PixelT	type,
		void *	ptr,
		std::function< void(void *)>	deleter
	)

Creates a new MRImage with dimensions set by ndim, and size set by size. Output pixel type is decided by type variable.

Parameters

dim	size of image, in each dimension, number of dimensions decied by length of size vector
type	Pixel type npl::PixelT
ptr	Pointer to data block
deleter	function to delete data block

Returns: New image, default orientation

ptr<MRImage> npl::diffOfGauss	(	ptr< const MRImage >	in,
		double	sd1,
		double	sd2
	)

Computes difference of gaussians.

Parameters

in	Gaussian smooths an image twice and subtracts
sd1	Standard deviation of first image
sd2	Standard deviation of second image

Returns: Difference of two different gaussians

ptr<MRImage> npl::fft_backward	(	ptr< const MRImage >	in,
		const std::vector< size_t > &	in_osize
	)

Performs inverse FFT transform in N dimensions.

Parameters

in	Input image
in_osize	Size of output image. If this is smaller than the input then the frequency domain will be trunkated, if it is larger then the fourier domain will be padded ( output upsampled )

Returns: Frequency domain of input. Note the output will be COMPLEX128/CDOUBLE type

ptr<MRImage> npl::fft_forward	(	ptr< const MRImage >	in,
		const std::vector< size_t > &	in_osize
	)

Performs forward FFT transform in N dimensions.

Parameters

in	Input image
in_osize	Size of output image (will be padded up to this prior to FFT)

Returns: Frequency domain of input. Note the output will be COMPLEX128/CDOUBLE type

void npl::gaussianSmooth1D	(	ptr< MRImage >	inout,
		size_t	dim,
		double	stddev
	)

Gaussian smooths an image in 1 direction. Questionable whether it works. Seems to shift image.

Parameters

inout	Input/Output image
dim	Direction to smooth in
stddev	in real space, for example millimeters.

std::ostream & npl::operator<<	(	std::ostream &	out,
		const MRImage &	img
	)

Writes out information about an MRImage.

Parameters

out	Output ostream
img	Image to write information about

Returns: More ostream

double npl::overlapRatio	(	ptr< const MRImage >	a,
		ptr< const MRImage >	b
	)

Computes the overlap of the two images' in 3-space.

Parameters

a	Image
b	Image

Returns: Ratio of b that overlaps with a's grid

ptr<MRImage> npl::randImage	(	PixelT	type,
		double	mean,
		double	sd,
		size_t	x,
		size_t	y,
		size_t	z,
		size_t	t
	)

Create random image, with gaussian distribution.

Parameters

type	Type of pixels
mean	mean of Gaussian
sd	Standard deviation of distribution
x	X size
y	Y size
z	Z size
t	Time size, default 0 time

Returns: Output MRImage

ptr<MRImage> npl::resample	(	ptr< const MRImage >	in,
		double *	spacing,
		double(*)(double, double)	window = `hannWindow`
	)

Performs fourier resampling using fourier transform and the provided window function.

Parameters

in	Input image
spacing	Desired output spacing
window	Window function to reduce ringing

Returns: Smoothed and downsampled image

ptr<MRImage> npl::resampleNN	(	ptr< const MRImage >	in,
		ptr< const MRImage >	atlas,
		PixelT	type = `UNKNOWN_TYPE`
	)

Performs nearest neighbor resasmpling of input to atlas.

Parameters

in	Input image
atlas
type	of output pixel (defaults to input type)

Returns: Input image resampled into atlas space

ptr<MRImage> npl::resampleNN	(	ptr< const MRImage >	in,
		double *	newspace,
		PixelT	type = `UNKNOWN_TYPE`
	)

Performs nearest neighbor resasmpling of input to atlas.

Parameters

in	Input image
newspace	new spacing
type	pixel type of output (defaults to input)

Returns: Input image resampled into atlas space

ptr<MRImage> npl::rigidTransform	(	ptr< MRImage >	in,
		double	rx,
		double	ry,
		double	rz,
		double	sx,
		double	sy,
		double	sz
	)

Rigid Transforms an image.

Parameters

in	Input/output image
rx	Rotation about x axis
ry	Rotation about y axis
rz	Rotation about z axis
sx	shift about x axis
sy	shift about y axis
sz	shift about z axis

int npl::rotateImageShearFFT	(	ptr< MRImage >	inout,
		double	rx,
		double	ry,
		double	rz,
		double(*)(double, double)	window = `npl::sincWindow`
	)

Rotates an image around the center using shear decomposition followed by FFT-based shearing. Rotation order is Rz, Ry, Rx, and about the center of the image.

Parameters

inout	Input/output image
rx	Rotation about x axis
ry	Rotation about y axis
rz	Rotation about z axis
rz	Rotation about z axis
window	Window function to apply in fourier domain

int npl::rotateImageShearKern	(	ptr< MRImage >	inout,
		double	rx,
		double	ry,
		double	rz,
		double(*)(double, double)	kern = `npl::lanczosKern`
	)

Rotates an image around the center using shear decomposition followed by kernel-based shearing. Rotation order is Rz, Ry, Rx, and about the center of the image. This means that 1D interpolation will be used.

Parameters

inout	Input/output image
rx	Rotation about x axis
ry	Rotation about y axis
rz	Rotation about z axis
kern	Kernel to use for interplation

ptr<MRImage> npl::shiftImage	(	ptr< MRImage >	in,
		size_t	len,
		double *	vect
	)

Uses fourier shift theorem to shift an image.

Parameters

in	Input image to shift
len	length of dx array
vect	movement in physical coordinates, will be rotated using image orientation prior to shifting

Returns: shifted image

ptr<MRImage> npl::smoothDownsample	(	ptr< const MRImage >	in,
		double	sigma,
		double	spacing = `-1`
	)

Performs smoothing in each dimension, then downsamples so that pixel spacing is roughly equal to FWHM.

Parameters

in	Input image
sigma	Standard deviation for smoothing
spacing	Ouptut image spacing (isotropic). If this is <= 0, then sigma will be used, which is a very conservative downsampling.

Returns: Smoothed and downsampled image

void npl::writeComplex	(	std::string	basename,
		ptr< const MRImage >	in,
		bool	absPhase = `false`
	)

Writes a pair of images, one real, one imaginary or if absPhase is set to true then an absolute image and a phase image.

Parameters

basename	Base filename _abs.nii.gz and _phase.nii.gz or _re.nii.gz and _im.nii.gz will be appended, depending on absPhase
in	Input image
absPhase	Whether the break up into absolute and phase rather than re/imaginary

Functions

Detailed Description

Function Documentation