A Hybrid Density-Based Clustering Pipeline for Track Reconstruction

In high-energy physics, the capability to accurately and efficiently track charged particles is essential for effective data analysis. This article introduces an innovative density-based clustering pipeline intended for the track reconstruction task, incorporating Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and Ordering Points To Identify the Clustering Structure (OPTICS) algorithm. Results on simulated data suggest that the proposed method offers improvements in both effectiveness and robustness compared to traditional techniques, with performance on par with state-of-the-art neural network-based approaches. Furthermore, this pipeline demonstrates significant potential for real-time applications in high-energy physics experiments, offering a scalable and robust solution.