Noise Module

Methods for adding artificial noise to data

@author: Antony Vamvakeros

nDTomo.methods.noise.addpnoise1D(sp, ct)

Add Poisson noise to a 1D spectrum.

Parameters:

• sp (ndarray) – 1D array representing the spectrum (e.g., intensity values).

• ct (float) – Scaling factor representing the total counts or acquisition time.

Returns:

Noisy spectrum with Poisson noise applied and rescaled by the count factor.

Return type:

ndarray

nDTomo.methods.noise.addpnoise2D(im, ct)

Add Poisson noise to a 2D image.

Parameters:

• im (ndarray) – 2D array representing the image (e.g., integrated intensity map).

• ct (float) – Scaling factor representing the total counts or acquisition time.

Returns:

Noisy image with Poisson noise applied and rescaled by the count factor.

Return type:

ndarray

nDTomo.methods.noise.addpnoise3D(vol, ct)

Adds Poisson noise to a 3D hyperspectral volume (H x W x Bands), noise is added per pixel-spectrum (i.e., per (i,j,:)).

Parameters:

• vol (ndarray) – 3D hyperspectral image (H x W x Bands), must be non-negative.

• ct (float) – Scaling constant to simulate photon counts.