Optical Transfer Functions#
These are functions for converting a 3D point spread function (PSF) volume into
a radially averaged 2D complex Optical Transfer Function (OTF) that can be used
for deconvolution. You can either write the OTF to file for later use using
make_otf function, or use the
TemporaryOTF context manager to create and delete a
temporary OTF from a 3D PSF input.
- class pycudadecon.TemporaryOTF(psf, **kwargs)#
Context manager to read OTF file or generate a temporary OTF from a PSF.
Normalizes the input PSF to always provide the path to an OTF file, converting the PSF to a temporary file if necessary.
self.pathcan be used within the context to get the filepath to the temporary OTF filepath.
- psf (str, np.ndarray): 3D PSF numpy array, or a filepath to a 3D PSF
or 2D complex OTF file.
- **kwargs: optional keyword arguments will be passed to the
OTF files cannot currently be provided directly as 2D complex np.ndarrays
ValueError: If the PSF/OTF is an unexpected type NotImplementedError: if the PSF/OTF is a complex 2D numpy array
>>> with TemporaryOTF(psf, **kwargs) as otf: ... print(otf.path)
- pycudadecon.make_otf(psf, outpath=None, dzpsf=0.1, dxpsf=0.1, wavelength=520, na=1.25, nimm=1.3, otf_bgrd=None, krmax=0, fixorigin=10, cleanup_otf=False, max_otf_size=60000, **kwargs)#
Generate a radially averaged OTF file from a PSF file.
psf (str) – Filepath of 3D PSF TIF
outpath (str, optional) – Destination filepath for the output OTF (default: appends “_otf.tif” to filename), by default None
dzpsf (float, optional) – Z-step size in microns, by default 0.1
dxpsf (float, optional) – XY-Pixel size in microns, by default 0.1
wavelength (int, optional) – Emission wavelength in nm, by default 520
na (float, optional) – Numerical Aperture, by default 1.25
nimm (float, optional) – Refractive indez of immersion medium, by default 1.3
otf_bgrd (int, optional) – Background to subtract. “None” = autodetect., by default None
krmax (int, optional) – pixels outside this limit will be zeroed (overwriting estimated value from NA and NIMM), by default 0
fixorigin (int, optional) – for all kz, extrapolate using pixels kr=1 to this pixel to get value for kr=0, by default 10
cleanup_otf (bool, optional) – clean-up outside OTF support, by default False
max_otf_size (int, optional) – make sure OTF is smaller than this many bytes. Deconvolution may fail if the OTF is larger than 60KB (default: 60000), by default 60000
Path to the OTF file
- Return type