A New Method of Open Cluster Membership Determination

Membership determination is the key-important step to study open clusters, which can directly influence on the estimation of open clusters’ physical parameters. DBSCAN (Density Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm in data mining techniques. In this paper the DBSCAN algorithm has been used for the first time to make the membership determination of the open clusters NGC 6791 and M 67 (NGC 2682). Our results indicate that the DBSCAN algorithm can effectively eliminate the contamination of field stars. The obtained member stars of NGC 6791 exhibit clearly a doubled main-sequence structure in the color-magnitude diagram, implying that NGC 6791 may have a more complicated history of star formation and evolution. The clustering analysis of M67 indicates the presence of mass segregation, and the distinct relative motion between the central part and the outer part of the cluster. These results demonstrate that the DBSCAN algorithm is an effective method of membership determination, and that it has some advantages superior to the conventional kinematic method.