Cosmic ray particles reaching the earth's surface from space can be used for imaging purposes

The physical processes that occur when a particle passes through a matter can be reconstructed, which can be used to study the structure of objects. Unlike e.g. Computer Tomography (CT) in which the radiation source is artificial, in muon tomography the reconstruction of the examined space must be supported by additional tools. In this case, AI/ML methods such as segmentation or object recognition greatly increase the quality of the results.

During the presentation, I will discuss the results of muon tomography based on muons from cosmic rays. I will present the simulation process that generates cosmic ray particles. I will also discuss the methods for identifying the interactions of muons with matter. The latest imaging results using segmentation and material classification will be reported.

This study is a part of the machine learning competition published on the Knowledge Pit platform as a part of the IEEE Big Data 2023 conference, and is also the subject of my doctoral thesis.