Paper ID: 2304.14078

ganX -- generate artificially new XRF a python library to generate MA-XRF raw data out of RGB images

Alessandro Bombini

In this paper we present the first version of ganX -- generate artificially new XRF, a Python library to generate X-ray fluorescence Macro maps (MA-XRF) from a coloured RGB image. To do that, a Monte Carlo method is used, where each MA-XRF pixel signal is sampled out of an XRF signal probability function. Such probability function is computed using a database of couples (pigment characteristic XRF signal, RGB), by a weighted sum of such pigment XRF signal by proximity of the image RGB to the pigment characteristic RGB. The library is released to PyPi and the code is available open source on GitHub.

Submitted: Apr 27, 2023