The Cognitive Approach in Cloud Computing and Internet of Things Technologies for Surveillance Tracking Systems

Dinesh Peter, Amir Hossein Alavi, Bahman Javadi, Steven L. Fernandes

• Publisher: Academic Press
• Year: 2020
• ISBN:
  9780128166093
• Topic: Scienze biologiche
• Subtopic: Biotecnologia

The Cognitive Approach in Cloud Computing and Internet of Things Technologies for Surveillance Tracking Systems discusses the recent, rapid development of Internet of things (IoT) and its focus on research in smart cities, especially on surveillance tracking systems in which computing devices are widely distributed and huge amounts of dynamic real-time data are collected and processed. Efficient surveillance tracking systems in the Big Data era require the capability of quickly abstracting useful information from the increasing amounts of data. Real-time information fusion is imperative and part of the challenge to mission critical surveillance tasks for various applications.

This book presents all of these concepts, with a goal of creating automated IT systems that are capable of resolving problems without demanding human aid.

• Examines the current state of surveillance tracking systems, cognitive cloud architecture for resolving critical issues in surveillance tracking systems, and research opportunities in cognitive computing for surveillance tracking systems
• Discusses topics including cognitive computing architectures and approaches, cognitive computing and neural networks, complex analytics and machine learning, design of a symbiotic agent for recognizing real space in ubiquitous environments, and more
• Covers supervised regression and classification methods, clustering and dimensionality reduction methods, model development for machine learning applications, intelligent machines and deep learning networks
• includes coverage of cognitive computing models for scalable environments, privacy and security aspects of surveillance tracking systems, strategies and experiences in cloud architecture and service platform design

Chapter 1

Reliable Surveillance Tracking System based on Software Defined Internet of Things

Deva Priya Isravel, Salaja Silas and Elijah Blessing Rajsingh, Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Coimbatore, India

Abstract

The creation of digital society has enabled individuals to have access to information from anywhere and at anytime across the Internet. The advancement in technologies like the Internet of Things (IoT), cloud computing, and BigData supports a variety of small- to large-scale applications. The surveillance tracking system is one such application, where multiple surveillance devices become part of the network to observe and detect unusual events in a particular area. Despite widespread adoption, modern surveillance network is in need of new paradigm technology for enhancing its performance. Software defined networking (SDN) is a new technology that offers agility, programmability, and flexibility to the network operation. In this paper, the significance, classification, applications, and challenges of the surveillance tracking system are presented. The different communication technologies used in the surveillance tracking system are discussed. In an effort to improve the current surveillance tracking system, the SDN-assisted IoT solution is provided and proposed a novel SDN-based traffic engineering framework for performance enhancement.

Keywords

Surveillance tracking system; software defined networking; Internet of Things; traffic engineering

Chapter Outline

Outline

• 1.1 Introduction 1
• 1.2 Surveillance Tracking System 2
  • 1.2.1 Classification of the Surveillance 3
  • 1.2.2 Applications 4
  • 1.2.3 Challenges 6
• 1.3 Wireless Communication Technologies 6
• 1.4 Software Defined Networking 8
• 1.5 Software Defined Surveillance Tracking System 9
  • 1.5.1 Traffic Engineering 10
  • 1.5.2 Proposed Traffic Engineering Framework 11
• 1.6 Conclusion 14
• References 14

1.1 Introduction

Intelligent surveillance tracking system provides real-time and sustained monitoring of a person, groups of people, objects, behavior, events, or environment. In recent years, there has been a rise in the use of surveillance tracking system for numerous applications. They are widely used in military applications, public monitoring, and commercial purposes. The main purpose of these kinds of observation is to provide personal and public safety, identify crime, prevent criminal activity, and enhance businesses and scientific research. Surveillance is extensively used for monitoring the safety of people from street corners to crowded places such as railways, airports, restaurants, malls, etc. It is also widely used in health-care services for observing patients and hospital facilities to provide quality care and support emergency preparedness and emergency services [1,2]. A lot of businesses use surveillance to boost their company productivity and profit by monitoring less their employees and concentrating more on businesses.

The benefits of a surveillance system can be applied in a number of ways other than security purpose. They are applied for building smart home automation and smart city projects [3–5]. The recent interest in mass surveillance for various causes introduces increased complexity in managing the surveillance system. With the recent advancement of new disruptive technologies, reliable and sophisticated surveillance tracking is built with multiple features. The surveillance tracking system should be able to provide a fast, time-sensitive, reliable, and rapid recovery mechanism for monitoring and predicting possible dangerous situations. The scale and complexity of surveillance networks are approaching massive and rapid changes. With the rise in the proportion of Internet of Things (IoT) enabled devices, sensors, mobile devices, smartphones, etc., the total Internet traffic has grown tremendously. A number of devices communicating at the same time with the base station increase and congestion occurs in the surveillance network. The amount of traffic exchanged across devices is also huge. Managing huge volumes of data traffic generated from multiple monitoring and capturing devices is complex, because it has to be processed simultaneously and sent to the appropriate base station or to the cloud for further investigation and data analytics. Because of this, the traditional network architecture has huge complexity and challenges in handling the network traffic and network management. Therefore improvising the intelligent and automated surveillance tracking system requires the scientific and research community to provide solutions. To address the challenges faced by traffic management, a new software defined networking (SDN) technology can be integrated into the surveillance tracking system to enhance the data transmission concerns that exist in the legacy surveillance network.

Section 1.2 presents in detail the concepts of the surveillance tracking system. Section 1.3 discusses the various communication technologies that are already in use to deploy the surveillance system. Section 1.4 provides a brief overview of the SDN technologies and its benefits when combined with the IoT. The novel framework of SDN-assisted IoT solution for building an effective reliable surveillance system is discussed in Section 1.5. Finally, Section 1.6 provides the conclusion.

1.2 Surveillance Tracking System

The surveillance tracking system is a system that is used for tracking humans, objects, vehicles, etc. and monitoring environment for ensuring safety and avoiding intruders. The surveillance has become a necessity for monitoring public and private spaces. Modern surveillance systems have demanding requirements with enormous, busy, and complex scenes, with heterogeneous sensor networks. The real-time acquisition and interpretation of the environment and flagging potentially critical situations are challenging [6]. The implementation of the surveillance system has three major phases. They are data capturing, data analysis, and postprocessing. In the data capturing phase, the web traffic, audio/video, and VoIP contents are captured from the environment and given as input to the preprocessing module for extraction. The data analysis phase comprises different steps in processing to obtain an enhanced quality image. The steps are image preprocessing, object-based analysis, event-based analysis, and visualization. In image preprocessing, video frames are extracted from the captured visual. Then interframes are estimated and image encoding is applied. The subregions of the image are identified by segmenting the image into partitions of different configurations in order to detect the person. The second phase of object-based analysis involves person tracking, posture classification, and body-based analysis. Then, estimations are then updated. The event-based analysis contains interaction modeling and activity analysis to explore the events happening. Finally, in the visualization stage, based on the camera calibration, an enhanced quality image is obtained. After the analysis phase, the extracted image is sent for postprocessing to take further evaluations and generated actions. Fig. 1.1 depicts the three major phase of the surveillance system.

Figure 1.1 Phases of surveillance tracking system.

1.2.1 Classification of the Surveillance

The surveillance tracking system can be broadly classified into three types. They are audio surveillance, video surveillance, and Internet surveillance.

1.2.1.1 Audio surveillance

Audio surveillance involves listening to sounds and detecting various acoustic events. Audio surveillance is applied to a wide range of applications like spying, patrolling, detective operations, etc. A number of sophisticated devices are available to work under different circumstances. Some of the listening devices are telephones, microphones, smartphones, wiretapping, voice recorders, and acoustic sensors. These devices capture the sound and then are analyzed to detect unusual and unsafe events [7]. The two important things in audio surveillance are feature extraction and audio pattern recognition [8].

1.2.1.2 Video surveillance

Video surveillance is monitoring the behavior or activity in an area by capturing video images, and these images are transferred to the automated system for further processing. The devices used are cameras, sensors, high definition capability video capturing devices, and display monitors to view the captured video in real time. Earlier days, the video surveillance system used simple video acquisition and display systems. But with the advancement in technologies, modern video surveillance tracking system has sophisticated devices for image and video acquisition and data processing. It can integrate image and video analysis algorithms for pattern recognition, decision-making, and image enhancement. The major task in video surveillance is the detection and recognition of moving objects, tracking, performing behavioral analysis, and retrieving of the important data of concern [9]. Extracting visual from long footages is a laborious and time-consuming task. Therefore visual analytics is required to process visual content without human intervention. Various tools and technologies are integrated to understand the different dimensions of video summarization, visualization, interact...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• List of Contributors
• Chapter 1. Reliable Surveillance Tracking System based on Software Defined Internet of Things
• Chapter 2. An Efficient Provably Secure Identity-Based Authenticated Key Agreement Scheme for Intervehicular Ad Hoc Networks
• Chapter 3. Dynamic Self-Aware Task Assignment Algorithm for an Internet of Things-Based Wireless Surveillance System
• Chapter 4. Smart Vehicle Monitoring and Tracking System Powered by Active Radio Frequency Identification and Internet of Things
• Chapter 5. An Efficient Framework for Object Tracking in Video Surveillance
• Chapter 6. Development of Efficient Swarm Intelligence Algorithm for Simulating Two-Dimensional Orthomosaic for Terrain Mapping Using Cooperative Unmanned Aerial Vehicles
• Chapter 7. Trends of Sound Event Recognition in Audio Surveillance: A Recent Review and Study
• Chapter 8. Object Classification of Remote Sensing Image Using Deep Convolutional Neural Network
• Chapter 9. Compressive Sensing-Aided Collision Avoidance System
• Chapter 10. Review of Intellectual Video Surveillance Through Internet of Things
• Chapter 11. Violence Detection in Automated Video Surveillance: Recent Trends and Comparative Studies
• Chapter 12. FPGA-Based Detection and Tracking System for Surveillance Camera
• Index