Layered software architecture for the development of third-generation video surveillance systems

Abstract

Mobile distributed systems of third-generation video surveillance (MDSV) have become a useful tool to provide multiple security services to people. For this type of systems, three key aspects must be carried out: 1) protection, which consists of preventing undesirable events; 2) detection, which refers to determining the exact moment in which the event occurred; and 3) reply, in this regard, actions such as activating alarms and generating warnings are executed. Previous works have proposed software architectures to development video surveillance systems on mobile distributed systems (MDS). However, these architectures focus mainly on providing services/aspects of protection and detection; without considering in its design the requirements that arise from the characteristics of the MDS, such as limited processing and storage capacities of devices, frequent disconnections, among others. In this paper, we introduce a layered software architecture to build MDSV. The proposed architecture considers and satisfies the requirements that arise from the critical aspects of protection, detection, and reply, including the characteristics of the MDS. Based on our architecture, an MDSV prototype was implemented. The tests carried out on the prototype show that the proposed architecture correctly provides users with various essential services in terms of protection, detection, and reply. From our point of view, the most important advantages of our proposed software architecture are the following: define the basic technical guidelines that an MDSV must have and accomplish; streamline overall development, providing a solid framework for developers; and contribute to satisfying the requirements that arise from quality attributes that the MDSV must possess.