A Location-based Architecture for Video Stream Selection in the Context of IoMT

Location-Based Services (LBS) are part of our day-to-day life, assisting in a variety of tasks such as tracing better routes to a destination, traffic verification and even for safety applications such as car or people tracking. However, it may not possible visually to identify some activities like attempted theft or kidnapping in the same way that a surveillance camera with visual information. It was identified that the integration of these two information (Global Position and Video) could improve surveillance systems from a security perspective. IP cameras are common examples of multimedia sensors that have become an increasingly popular and ubiquitously installed item in urban centers applied to security. This paper proposes a location-based architecture for video stream selection that can be used to monitor entities (people or not) in a context of Internet of Multimedia Things (IoMT) through a representational model of coverage areas of IP cameras. A prototype was implemented as a proof of concept of the proposed architecture. This prototype includes a low latency video transmission service to quickly start a video transmission only when an entity is inside a visual field of view. Tests were performed on the Internet to check for video delivery latency, bandwidth consumption, and the results suggest the effectiveness of the proposal.