All Pervasive Surveillance Techniques and AI-Based Applications: Current Trends and Challenges

Introduction

Over the years, a number of surveillance systems have been developed that have helped in managing security task, limiting the chances of intrusion or unauthorized access. The act of surveillance conducted for various reasons by different individuals, organisations, institutions etc. can be covert or overt depending on the requirement and intensity of applications. Security being the main issue today, protecting the life and property of the human has become a prominent matter of concern of the world today. The concept of security from manual door lock and key mechanism has now transformed into using more sophisticated security devices such as cameras, motion detection sensors, proximity sensors, alarms etc. In security surveillance system while maintaining the redundancy, the concept of total security is unavoidable such that the system continuously tracks the identity, location and activity of objects within the monitored space. A surveillance system typically focuses on tracking location and activity, while biometrics systems focuses on identifying individuals (Hampapur, Brown, Connell, Pankanti, Senior and Tian, 2004). The ability to recognise objects and humans, to describe their actions and interactions from information acquired by sensors is essential for automated visual surveillance (Valera and Velastin, 2004). The increasing need for intelligent visual surveillance in commercial, law enforcement and military applications makes automated visual surveillance systems one of the main current application domains in computer vision (Valera and Velastin, 2004).

With the advent of technologies and services, the facilities of interaction between human and machine has increased to a great extent. It is an often observed phenomenon of increasing the use of innovative security-aids with the rise in information and communication technology (ICT) and proliferation of human habitation. The user friendly nature and the facility to access to information has made automation a significant part of ICT systems. The automation has triggered many changes in the existing technologies. The security systems are designed taking into consideration the safety of the person or objects. Thus, providing security lies in factors such as equipment or technology and people. Taking into account the various issues related to security such as ease of access, lack of intrusion detection alert, inefficient monitoring method (Ansari, Sedkyl, Sharma and Tyagi, 2015), the different methods employed in monitoring is the key to prevent dangerous situations. This involves resorting an automated approach of monitoring and controlling once deployed. The ability of the network devices to connect to various other devices involves collection of a large amount of data from around the world and sharing that data over the internet, a host of application has been evolving, making revolution in technology. A part of such a technology and mechanism lies in continuous monitoring of the sensitive areas like home or areas such as an office, institutions, traffic point, airport entrance, defence installations, hospitals, roadsides, buildings, elevators, commercial organisations etc. which involves real-time monitoring, capturing, hundreds and thousands of images each day. It has become a prominent part of the world today to make security an important aspect for the safety of life and property of person. The world is now entering into the era of Artificial intelligence (AI) and Internet of Things (IoT) which has been evolving at a great extent. This rise in the evolution of new connected devices and learning mechanism has brought about fundamental changes to the society. IoT which is a new leap in technology of network and connected device started with work on radio RFID tags and has expanded to connected sensors and actuators, along with many other technologies designed for different purposes (e.g., ZigBee, NFC, Bluetooth) including communication technology. With its impact in larger domain in the near future (Kolias, Voas, Bojanova and Kuhn, 2016) IoT devices will become the mainstream ranging from healthcare to retail to transportation. From the literature it has been found that a considerable amount of work has been done in surveillance systems over the past decades. The paper reports the work related to the integration of devices for physical intrusion detection and throws light on the means of increasing the level of security in sensitive areas.