def nufft_adjoint(input, coord, oshape, oversamp=1.25, width=4.0, n=128, device='cuda'):
ndim = coord.shape[-1]
beta = numpy.pi * (((width / oversamp) * (oversamp - 0.5)) ** 2 - 0.8) ** 0.5
oshape = list(oshape)
os_shape = _get_oversamp_shape(oshape, ndim, oversamp)
# Gridding
coord = _scale_coord(coord, oshape, oversamp, device)
kernel = _get_kaiser_bessel_kernel(n, width, beta, coord.dtype, device)
output = interp.gridding(input, os_shape, width, kernel, coord, device)
# IFFT
output = output.permute(0, 2, 3, 1)
# plt.figure()
# plt.imshow(print_complex_kspace_tensor(output[0].detach().cpu()), cmap='gray')
# plt.show()
output = transforms.ifft2(output)
# plt.figure()
# plt.imshow(print_complex_image_tensor(output[0].detach().cpu()), cmap='gray')
# plt.show()
# Crop
output = output.permute(0, 3, 1, 2)
output = util.resize(output, oshape, device=device)
output *= util.prod(os_shape[-ndim:]) / util.prod(oshape[-ndim:]) ** 0.5
# Apodize
output = _apodize(output, ndim, oversamp, width, beta, device)
return output.permute(0, 2, 3, 1)
def nufft_adjoint(input, coord, out_shape, oversamp=1.25, width=4.0, n=128, device='cuda'):
ndim = coord.shape[-1]
beta = numpy.pi * (((width / oversamp) * (oversamp - 0.5)) ** 2 - 0.8) ** 0.5
out_shape = list(out_shape)
os_shape = _get_oversamp_shape(out_shape, ndim, oversamp)
# Gridding
out_shape2 = out_shape.copy()
os_shape2 = os_shape.copy()
coord = _scale_coord(coord, out_shape2, oversamp, device)
kernel = _get_kaiser_bessel_kernel(n, width, beta, coord.dtype, device)
output = interp.gridding(input, os_shape2, width, kernel, coord, device)
# IFFT
output = output.permute(0, 1, 3, 4, 2)
output = transforms.ifft2_regular(output)
output = output.permute(0, 1, 4, 2, 3)
# Crop
output = util.resize(output, out_shape2, device=device)
a = util.prod(os_shape2[-ndim:]) / util.prod(out_shape2[-ndim:]) ** 0.5
output = output * a
# Apodize
output = _apodize(output, ndim, oversamp, width, beta, device)
return output
def gridding(input, shape, width, kernel, coord, device):
ndim = coord.shape[-1]
batch_shape = shape[:-ndim]
batch_size = util.prod(batch_shape)
pts_shape = coord.shape[:-1]
npts = util.prod(pts_shape)
input = input.reshape([batch_size, npts])
coord = coord.reshape([npts, ndim])
output = torch.zeros([batch_size] + list(shape[-ndim:]), dtype=input.dtype, device=device)
output = _gridding2(output, input, width, kernel, coord)
return output.reshape(shape)
def interpolate(input, width, kernel, coord, device):
ndim = coord.shape[-1]
batch_shape = input.shape[:-ndim]
batch_size = util.prod(batch_shape)
pts_shape = coord.shape[:-1]
npts = util.prod(pts_shape)
input = input.reshape([batch_size] + list(input.shape[-ndim:]))
coord = coord.reshape([npts, ndim])
output = torch.zeros([batch_size, npts], dtype=input.dtype, device=device)
output = _interpolate2(output, input, width, kernel, coord)
return output.reshape(batch_shape + pts_shape)
def nufft(input,
coord,
ndim=2,
oversamp=1.25,
width=4.0,
n=128,
device='cuda'):
# ndim = coord.shape[-1]
beta = numpy.pi * (((width / oversamp) * (oversamp - 0.5))**2 - 0.8)**0.5
os_shape = _get_oversamp_shape(input.shape, ndim, oversamp)
output = input.clone()
# Apodize
output = _apodize(output, ndim, oversamp, width, beta, device)
# Zero-pad
output = output / util.prod(input.shape[-ndim:])**0.5
output = util.resize(output, os_shape, device=device)
# FFT
if ndim == 2:
output = transforms.rfft2(output)
elif ndim == 3:
output = transforms.rfft3(output)
# Interpolate
coord = _scale_coord(coord, input.shape, oversamp, device)
kernel = _get_kaiser_bessel_kernel(n, width, beta, coord.dtype, device)
output = interp.interpolate(output, width, kernel, coord, ndim, device)
return output
def nufft_adjoint(input,
coord,
oshape,
ndim=2,
oversamp=1.25,
width=4.0,
n=128,
device='cuda'):
# ndim = coord.shape[-1]
beta = numpy.pi * (((width / oversamp) * (oversamp - 0.5))**2 - 0.8)**0.5
oshape = list(oshape)
os_shape = _get_oversamp_shape(oshape, ndim, oversamp)
# Gridding
oshape2 = oshape.copy()
oshape2[1] = 2
os_shape2 = os_shape.copy()
os_shape2[1] = 2
coord = _scale_coord(coord, oshape2, oversamp, device)
kernel = _get_kaiser_bessel_kernel(n, width, beta, coord.dtype, device)
output = interp.gridding(input, os_shape2, width, kernel, coord, ndim,
device)
# IFFT
if ndim == 2:
output = transforms.ifft2(output)
elif ndim == 3:
output = transforms.ifft3(output)
# Crop
output = util.resize(output, oshape2, device=device)
a = util.prod(os_shape[-ndim:]) / util.prod(oshape[-ndim:])**0.5
output = output * a
# Apodize
output = _apodize(output, ndim, oversamp, width, beta, device)
if ndim == 2:
output = output[:, 0, :, :]
elif ndim == 3:
output = output[:, 0, :, :, :]
return output