VPEDA: A Robust Vantage Point Based Event Detection Architecture forEfficient Object Detection and Localization

摘要

Currently the identification and localization of mobile objects have come as one of the most important approaches for event detection on road. VANET applications works on the principle of local gradient method that evaluates the fitness between projection of the vehicle model and object data. However, the mobile objects are highly susceptible to the appearance of the vehicle and displacement that changes at a frequent time period. In VANET, each mobile object broadcasts its unique id, the appearance and displacement in beacon packets. The appearance and displacement are the two important factors to be considered during event detection. For that purpose, the traffic scenarios in road network have to be monitored at different perspectives to notice the location information of mobile objects. This study put forwards a Vantage Point-based Event Driven Architecture (VPEDA) as a novel method to get insight into different levels of events during different time periods. Furthermore, Object identification and position forecast using local gradient method considers only the exterior of the vehicle and displacement in the road network and does not discuss about the occurrences of event at the vehicle junction points. The proposed VPEDA focuses on the event detection using the different neighbourhood information of the mobile objects during diversified time interval using a vantage point. VPEDA process the occurrences of event like accident met at traffic junctions during different interval of time. The located event at a particular vantage point identifies the neighbour mobile object that is taken as the nodal point for standardization of the traffic modality. Moreover, with the nodal point, the locator ensures that the reason behind the event to be identified in an early stage. An investigational assessment is carried out to analyse the impact of the vantage point on average vehicle speed, percentage of packets delivered and. Experimental results shows that VPEDA can detect the event which involve the different number of mobile objects with minimum delay.