Detection of new gamma-ray AGN candidates using SIMEFIC III and DBSCAN in Fermi-LAT data.

We present the detection of new high-energy gamma-ray active galactic nucleus candidates using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm on two-dimensional gamma-ray data from Fermi Large Area Telescope. Our approach involves iterative applications of the SIeving MEthod for FInding Core (SIMEFIC III) algorithm, designed to enhance point source detection. By integrating DBSCAN at each denoising step, we track and evaluate the significance of potential sources across iterations. Our findings indicate that source significance increases to a specific threshold, beyond which further denoising may remove genuine sources. At the optimal denoising stage, we identified 18 sources not listed in the Fermi catalogue, with several of these sources potentially matching entries in the the Candidate Gamma-Ray Blazar Survey (CRATES) and Roma-BZCAT (Roma-BZCAT Multi-Frequency Catalog of Blazars) catalogues. [ABSTRACT FROM AUTHOR]