Skip to content

maurosilber/pawflim

Repository files navigation

pawFLIM: denoising via adaptive binning for FLIM datasets

PyPi PyPi License Paper

Installation

pawFLIM can be installed from PyPI:

pip install pawflim

or conda-forge:

conda install -c conda-forge pawflim

Usage

import numpy as np
from pawflim import pawflim

data = np.empty((3, *shape), dtype=complex)
data[0] = ...  # number of photons
data[1] = ...  # n-th (conjugated) Fourier coefficient
data[2] = ...  # 2n-th (conjugated) Fourier coefficient

denoised = pawflim(data, n_sigmas=2)

phasor = denoised[1] / denoised[0]

Note that we use the standard FLIM definition for the $n$-th phasor $r$:

$$ r_n = \frac{R_n}{R_0} $$

where

$$ R_n = \int I(t) , e^{i n \omega t} dt $$

is the $n$-th (conjugated) Fourier coefficient.

See the notebook in examples for an example with simulated data.