Cloud Architecture for IoT-Based Monitoring Services Using Query by Shape Object Identification

Usually in existing cities there are many different surveillance systems which are deployed inside and outside the buildings, for instance monitoring systems within private companies, inside households, in public institutions and also on the streets throughout the whole city. These systems function independently and in their traditional form they are supervised only by a human operator, who may potentially fail to notice some important events recorded by stationary cameras. In this paper we present a cloud architecture for monitoring services which allows the integration of these systems. Our approach is consistent with the idea of Internet of Things and with the concept of Smart City. Because reliability of a monitoring system is very important, computer vision algorithms can be used for automated detection of objects in a video signal. Our goal is to build the real-time system efficient at robust identification of objects on the basis of their approximate shape, which can also be used as a web service in a cloud. Object identification task is performed using our Query by Shape (QS) method.