vamtoolbox.util.data

Module Contents

Functions

sigmoid(x, g)
clipToCircle(x)	Sets all data outside the inscribed circle to zero
filterTargetOSMO(x, filter_name)	param x:
filterTargetBCLP(real_space_array, filter_name)	param real_space_array:
filterTarget(real_space_array, filter_name)	param real_space_array:
filterSinogram(sinogram, filter_name)	Filters a set of sinogram for a 2D or 3D target
_get_fourier_filter(size, filter_name)	Construct the Fourier filter.
histogramEqualization(x, bit_depth[, output_dtype])
discretize(x, bit_depth, range[, output_dtype])	Digitizes a variable to requested bit depth and output data type

Attributes

fftmodule
fft

vamtoolbox.util.data.fftmodule

vamtoolbox.util.data.fft

vamtoolbox.util.data.sigmoid(x, g)

vamtoolbox.util.data.clipToCircle(x: numpy.ndarray)

Sets all data outside the inscribed circle to zero

Parameters:

x (np.ndarray) – square array to be modified

Returns:

input array with all data outside the inscribed circle set to zero

Return type:

x

vamtoolbox.util.data.filterTargetOSMO(x: numpy.ndarray, filter_name: str)

Parameters:

• x (np.ndarray) –

• filter_name (str) – type of filter to apply to target, options: “ram-lak”, “shepp-logan”, “cosine”, “hamming”, “hanning”, “none”

Returns:

x_filtered – direct output of filtering in frequency space

Return type:

np.ndarray

vamtoolbox.util.data.filterTargetBCLP(real_space_array: numpy.ndarray, filter_name: str)

Parameters:

• real_space_array (np.ndarray) –

• filter_name (str) – type of filter to apply to target, options: “ram-lak”, “shepp-logan”, “cosine”, “hamming”, “hanning”, “none”

Returns:

x_filtered – direct output of filtering in frequency space

Return type:

np.ndarray

vamtoolbox.util.data.filterTarget(real_space_array: numpy.ndarray, filter_name: str)

Parameters:

• real_space_array (np.ndarray) –

• filter_name (str) – type of filter to apply to target, options: “ram-lak”, “shepp-logan”, “cosine”, “hamming”, “hanning”, “none”

Returns:

x_filtered – direct output of filtering in frequency space

Return type:

np.ndarray

vamtoolbox.util.data.filterSinogram(sinogram: numpy.ndarray, filter_name: str)

Filters a set of sinogram for a 2D or 3D target

Parameters:

• sinogram (np.ndarray) – input sinogram

• filter_name (str) – type of filter to apply to sinogram, options: “ram-lak”, “shepp-logan”, “cosine”, “hamming”, “hanning”, “ram-lak_freq”, “none”

Returns:

sinogram_filt – filtered sinogram

Return type:

np.ndarray

vamtoolbox.util.data._get_fourier_filter(size: int, filter_name: str)

Construct the Fourier filter. This computation lessens artifacts and removes a small bias as explained in [1], Chap 3. Equation 61.

Parameters:

• size (int) – filter size. Must be even.

• filter_name (str) – Filter used in frequency domain filtering. Filters available: ram-lak, shepp-logan, cosine, hamming, hanning. Assign None to use no filter.

Returns:

fourier_filter – The computed Fourier filter.

Return type:

np.ndarray

References

[1]

AC Kak, M Slaney, “Principles of Computerized Tomographic Imaging”, IEEE Press 1988.

vamtoolbox.util.data.histogramEqualization(x: numpy.ndarray, bit_depth: int, output_dtype: numpy.dtype = float)

vamtoolbox.util.data.discretize(x: numpy.ndarray, bit_depth: int, range: list, output_dtype: numpy.dtype = float)

Digitizes a variable to requested bit depth and output data type

Parameters:

• x (nd.ndarray) – array to digitize

• bit_depth (int) – bit depth of output, 2^bit_depth number of bins

• range (list) – [min,max] values to discretize within

• output_dtype (np.dtype (optional)) – data type of resulting digitized array