IoT-Based Surveillance Camera Distribution Using Triangle Geometry

• Main Authors: Almalkawi, Islam T., Al-Abed, Loiy, Al-Karaki, Jamal N., Zapata, Manel Guerrero
• Format: Conference Proceeding
• Language: English
• Subjects: Camera Field of View (FoV); Clustering algorithms; Costs; Heuristic algorithms; Internet of Things (IoT); Redundancy; Reinforcement learning; Simulation; Surveillance; Surveillance Systems; Triangle Geometry
• ISSN: 2576-3555
• DOI: 10.1109/CoDIT58514.2023.10284365

IoT systems provide excellent solutions to many applications in both civilian and military aspects. This is even facilitated by the advent of many types of cameras with the diminishing cost of communication devices. We focus on the surveillance system that collaboratively monitors an intended area. A good design of such systems should ensure complete coverage with the ability to detect and track objects. Existing surveillance systems produce sensing coverage redundancy with extra communication overhead and cost. This paper presents an IoT-based surveillance system that exhibits many features. First, it provides an efficient distributed network clustering algorithm based on the Field of View (FoV) of cameras. Second, maximum area coverage can be achieved with a minimum number of camera sensors. Simulation results show that our proposed camera distribution algorithm outperforms other existing techniques in terms of the needed number of cameras, coverage percentage, and cluster size. We also believe that the efficiency of our algorithm can be improved with the involvement of the reinforcement learning approach where dynamic camera adaptation can be achieved based on the traffic distribution and changing conditions in the surveillance area.