Discriminating Most Urgent Trajectories in a Road Network using Density Based Online Clustering

Moving object trajectory patterns are clustered based on similarity to discriminate abnormal activities. These objects are usually categorized as outliers. Emergency vehicles such as ambulance and fire engine may follow different paths, other than normal due to their urgency. The work done so far, categorize these objects as outlying trajectories and thus come under suspicious movement category. In this paper, we propose a method for outlier trajectory detection in a road network using online density based clustering and to categorize outlier trajectories as most urgent trajectory (MUT) and most suspicious trajectories (MST). Experimental results on synthetic MOD (Moving Object Database) verify the effectiveness of the proposed scheme.