Artificial Intelligence in Video Surveillance

Artificial Intelligence in Video Surveillance

Uma Maheswari P., Karishma V. R., T. Vigneswaran
DOI: 10.4018/978-1-6684-8098-4.ch001
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Surveillance is an essential component of security, and e-surveillance is one of the primary goals of the Indian Government's Digital India development initiative. Video surveillance offers a wide range of applications to reduce ecological and economic losses and becomes one of the most effective means of ensuring security. This chapter addresses the problem of how artificial intelligence is powering video surveillance. There is a significant research focus on video analytics but comparatively less effort has been taken for surveillance videos. However, there is little evidence that researchers have approached the issue of intelligent video surveillance in terms of suspicious action detection, crime scene description, face detection, crowd counting, and the like. Most AI-powered surveillance is based on deep neural networks and deep learning techniques using analysis of video frames as images. Consequently, this chapter aims to provide an overview and significance of how artificial intelligence techniques are employed in video surveillance and image processing.
Chapter Preview
Top

Introduction

Today, security cameras have become an integral part of everyday life for the sake of safety and security. Surveillance camera installations in the private and public sectors have increased significantly to monitor public activities. Security experts focus significantly on video surveillance to combat crime and avoid unpleasant situations that harm human civilization. However, personal and corporate security cannot be achieved simply by installing a surveillance camera. The surveillance system should be sufficiently assisted with Artificial intelligence to deliver security solutions that substantially prevent abnormalities. Artificial intelligence has significantly influenced society, whether it takes the shape of algorithms, machine learning models, robotics, or autonomous systems. Many marketed video surveillance systems have integrated Artificial Intelligence (AI)-powered video analytics technology as a method to make our lives smarter and safer, thanks to recent developments in deep learning technologies. Intelligent Visual Surveillance is a significant and hard area of image processing and computer vision research. As our society is rapidly evolving toward smart homes and smart cities, necessitating an increasing number of Internet of Things (IoT) device deployments.

Background

The application of artificial intelligence (AI) is becoming increasingly crucial in the quest for novel techniques and technologies. Clutter identification, target categorization, and target tracking are AI techniques for target surveillance with radar sensors. These are critical assets for effective target observation. Because clutter (i.e., unwanted signal reflections) may significantly hamper target detection, its identification and subsequent suppression are critical. Furthermore, accurate target classification can aid in the successful prevention of possible threats, particularly in military circumstances. Finally, target tracking, the final link in the traditional chain of radar data processing, demands special attention since it provides the pivot point for sensor data fusion.

Complete Chapter List

Search this Book:
Reset