The Safety Transformation in the Future Internet Domain

Abstract

Public Safety is nowadays a priority, cornerstone and major concern for governments, majors and policy makers in current (and future) smart cities. Notwithstanding the foregoing, large advances in ICT technologies are foretold to revolutionize our society and enhance our feeling of safety (and hopefully, wellbeing). This chapter presents an introduction to three of the most promising technological pillars considered to be spearheads in this transformation: Internet of things, understood as the data capillarity through billions of sensors, Intelligent Video Analytics and Data Mining Intelligence, the latter two enabling smarter contextual awareness and prediction of potential threats leading to proactive prevention of them. The associated horizontal economic implications of this evolution and its impact into the societal and economic fabric are also tackled. Part of the results and analysis produced in this chapter are the outcome of the work carried out in the FP7 EU project SafeCity, one of the eight Use Cases of the FI Programme. The Safety Transformation in the Future Internet Domain 191 to reduce emergency response time and urban crime: for example, digital surveillance cameras have been placed in many critical areas and buildings throughout cities and call dispatchers have been created to distribute the emergency calls. Moreover, advanced technological capabilities facilitate urban public safety systems to become not just more interconnected and efficient, but also smarter and self-adaptive. Instead of merely responding to crimes and emergencies after a critical situation, novel smart systems emerge to analyse, anticipate and, actually, contribute to preventing them before occurring. After the terrorist attacks of March 2004 in Madrid, the city developed a new fully integrated Emergency Response Centre which, after an incoming emergency call, simultaneously alerts the required emergency agency (police, ambulance and/or fire brigade). The system can recognize if alerts relate to a single or multiple incidents, and assign the right resources based on the requirements coming from the ground. Furthermore, specialized video analytics systems are successfully installed for traffic surveillance purposes. These are CCTV-based systems capable of automatically detect illegal vehicles behaviour (e.g. cars stopped in forbidden areas, going in the opposite direction), restricted entries behaviour (e.g. bike entering in a forbidden road), stolen vehicles, etc. In addition, M2M communications, that is, intelligent communications by enabled devices without human intervention, are nowadays present in home and industrial security monitoring systems and alarms. Several Public Safety organizations and Public Administrations are using sensor networks to monitor environmental conditions or to be temporally deployed driven by an emergency situation. Other advanced technologies are focused on enhancing emergency notification mechanisms, fire and enforcement records management, surveillance, etc. As presented, outstanding capabiliti...

Analytics for the Internet of Things

“…A human activity recogniser combines signal processing, SVM and Gaussian Mixture Models to distinguish activities and recommends activities using rule-based reasoning. Sn Visualisation H Public Safety Monitoring [66] Video Video Analytics R Smart Building Heating [157] Sn Anomaly Detection R Memory Augmentation [75] IoT Data Data mining H Wearable Lifestyle Monitor [138] Sn Anomaly Detection R…”

“…Additionally, there are public safety and military applications that apply video analytics in detecting movement, intruders or targets. The public safety use case is elaborated on by Gimenez et al [66] where they discuss how given the big data problem of having huge amounts of video footage, smart video analytics systems can proactively monitor, automatically recognise and bring to notice situations, flag out suspicious people, trigger alarms and lock down facilities through the recognition of patterns and directional motion, recognising faces and spotting potential problems by tracking, with multiple cameras, how people move in crowded scenes.…”

“…5.2.4 Video Analytics. Video Analytics (VA) is about the use of specialised software and hardware "to analyse captured video and automatically identify specific objects, events, behaviour or attitudes in video footage in real-time" [66].…”