This paper presents a new method to process the data recorded with muon telescopes. We have developed this processing method for the plastic scintillator-based hodoscopes located around the volcano La Soufri\`ere de Guadeloupe, in the French Lesser Antilles, in order to perform muon radiographies of the lava dome region, strongly impacted by the volcanic hydrothermal activity. Our method relies on particle trajectory reconstruction, performing a fit of the recorded hits in the impacted scintillator bars using a Random Sample Consensus algorithm. This algorithm is specifically built to discriminate outlier points, usually due to noise hits, in the data. Thus, it is expected to significantly improve the signal/noise separation in muon track hits and to obtain higher quality estimates of the particles' incident trajectories in our detectors. The first analysis of the RANSAC-reconstructed events offers promising results in terms of average density maps. To illustrate the performances of this algorithm, we provide angular resolution and reconstruction efficiency estimates using a GEANT4 simulation of a telescope equipped with four detection matrices. In addition, we also show preliminary results from open-sky data recorded with such telescope at La Soufri\`ere de Guadeloupe volcano.
Read full abstract