I will present the application of clustering algorithms (DBSCAN and DENCLUE) to the detection and the deblending of confused sources, both for Optical/CCD and gamma-ray astronomical images. I will also show how to use the features extracted from these algorithms with supervised ensemble classification algorithms.

This presentation is part of Minisymposium “MS9 - Innovative models and algorithms for astronomical imaging (2 parts)”
organized by: Silvia Tozza (INdAM/Dept. Mathematics, University of Rome “La Sapienza”) , Marco Castellano (INAF Osservatorio Astronomico di Roma) , Maurizio Falcone (Dipartimento di Matematica, Università di Roma "La Sapienza") , Adriano Fontana (INAF Osservatorio Astronomico di Roma) .

Authors:

Andrea Tramacere (ISDC Data Center for Astrophysics, University of Geneve)

Keywords:

clustering algorithms, deblending of source confusion, image segmentation, machine learning