Clustering is crucial for many computer vision applications such as robust tracking, object detection and segmentation. Object detection/recognition finds its application in drones, autonomous driving, and so on. This work presents a real-time clustering technique that takes advantage of the unique properties of event-based vision sensors. Thus, this approach redefines the well-known clustering method using asynchronous events instead of conventional frames. Clustering accuracy reducing the computational cost by 88% compared to the frame-based method. The clustering achieved a consistent number of clusters along time. Event-based algorithm has been used for this cluster detection and we evaluated our method on a Verilog platform. This cluster detection mechanism is reliable, having less complexity on comparing with other image processing techniques.

Event-based processing, VHDL, Abnormal event detection, Linear feedback shift register( LFSR), cluster detection, Object Tracking, Cluster Object Tracker (COT).