Intelligent Multimedia Surveillance

Current Trends and Research

herausgegeben von: Pradeep K. Atrey, Mohan S. Kankanhalli, Andrea Cavallaro

Verlag: Springer Berlin Heidelberg

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

Intelligent multimedia surveillance concerns the analysis of multiple sensing inputs including video and audio streams, radio-frequency identification (RFID), and depth data. These data are processed for the automated detection and tracking of people, vehicles, and other objects. The goal is to locate moving targets, to understand their behavior, and to detect suspicious or abnormal activities for crime prevention. Despite its benefits, there is societal apprehension regarding the use of such technology, so an important challenge in this research area is to balance public safety and privacy.

This edited book presents recent findings in the field of intelligent multimedia surveillance emerging from disciplines such as multimedia computing, computer vision, and artificial intelligence. It consists of nine chapters addressing intelligent video surveillance, video analysis of crowds, privacy issues in intelligent multimedia surveillance, RFID technology for localization of objects, object tracking using visual saliency information, estimating multiresolution depth using active stereo vision, and performance evaluation for video surveillance systems.

The book will be of value to researchers and practitioners working on related problems in security, multimedia, and artificial intelligence.

Inhaltsverzeichnis

Frontmatter

Intelligent Video Surveillance as a Service

Abstract

Nowadays, intelligent video surveillance has become an essential tool of the greatest importance for several security-related applications. With the growth of installed cameras and the increasing complexity of required algorithms, in-house self-contained video surveillance systems become a chimera for most institutions and (small) companies. The paradigm of Video Surveillance as a Service (VSaaS) helps distributing not only storage space in the cloud (necessary for handling large amounts of video data), but also infrastructures and computational power. This chapter will briefly introduce the motivations and the main characteristics of a VSaaS system, providing a case study where research-lab computer vision algorithms are integrated in a VSaaS platform. The lessons learnt and some future directions on this topic will be also highlighted.

Andrea Prati, Roberto Vezzani, Michele Fornaciari, Rita Cucchiara

A Literature Review on Video Analytics of Crowded Scenes

Abstract

This chapter presents a review and systematic comparison of the state of the art on crowd video analysis. The rationale of our review is justified by a recent increase in intelligent video surveillance algorithms capable of analysing automatically visual streams of very crowded and cluttered scenes, such as those of airport concourses, railway stations, shopping malls and the like. Since the safety and security of potentially very crowded public spaces have become a priority, computer vision researchers have focused their research on intelligent solutions. The aim of this chapter is to propose a critical review of existing literature pertaining to the automatic analysis of complex and crowded scenes. The literature is divided into two broad categories: the macroscopic and the microscopic modelling approach. The effort is meant to provide a reference point for all computer vision practitioners currently working on crowd analysis. We discuss the merits and weaknesses of various approaches for each topic and provide a recommendation on how existing methods can be improved.

Myo Thida, Yoke Leng Yong, Pau Climent-Pérez, How-lung Eng, Paolo Remagnino

Privacy and Security in Video Surveillance

Abstract

Video surveillance systems are usually installed to increase the safety and security of people or property in the monitored areas. Typical threat scenarios are robbery, vandalism, shoplifting or terrorism. Other application scenarios are more intimate and private such as home monitoring or assisted living. For a long time, it was accepted that the potential benefits of video surveillance go hand in hand with a loss of personal privacy. However, with the on-board processing capabilities of modern embedded systems it becomes possible to compensate this privacy loss by making security and privacy protection inherent features of video surveillance cameras. In the first part of this chapter, we motivate the need for the integration of security and privacy features, we discuss fundamental requirements and provide a comprehensive review of the state of the art. The second part presents the TrustCAM prototype system where a dedicated hardware security module is integrated into a camera system to achieve a high level of security. The chapter is concluded by a summary of open research issues and an outlook to future trends.

Thomas Winkler, Bernhard Rinner

Object Video Streams: A Framework for Preserving Privacy in Video Surveillance

Abstract

Here we introduce a framework for preserving privacy in video surveillance. Raw video footage is decomposed into a background and one or more object-video streams. Such object-centric decomposition of the incoming video footage opens up new possibilities to provide visual surveillance of an area without compromising the privacy of the individuals present in that area. Object-video streams allow us to render the scene in a variety of ways: (1) individuals in the scene can be represented as blobs, obscuring their identities; (2) foreground objects can be color coded to convey subtle scene information to the operator, again without revealing the identities of the individuals present in the scene; (3) the scene can be partially rendered, that is, revealing the identities of some individuals, while preserving the anonymity of others, etc. We evaluate our approach in a virtual train station environment populated by autonomous, lifelike virtual pedestrians. We also demonstrate our approach on real video footage. Lastly, we show that Microsoft Kinect sensor can be used to decompose the incoming video footage into object-video streams.

Faisal Z. Qureshi

Surveillance Privacy Protection

Abstract

Surveillance Privacy Protection (SPP) is a realistic issue in the world we are living in today. Due to the massive progress in technologies and systems, surveillance is becoming quite impossible to avoid. More information is being handed out without realizing the risks involved. The objective of this chapter is to evaluate what types of surveillance, privacy and protection measures are being implemented, how information is being used and what rights individuals have over this. In addition, this chapter also emphasizes the importance of tools, data sets and databases that are being developed to enable surveillance privacy.

Nikki Gulzar, Basra Abbasi, Eddie Wu, Anil Ozbal, WeiQi Yan

RFID Localization Improved by Motion Segmentation in Multimedia Surveillance Systems

Abstract

An important issue in multimedia surveillance systems is determining the physical location of moving objects. Due to features like contactless communications, high data rate, non-line-of-sight readability, compactness and low cost, passive Radio Frequency Identification technology is very attractive for indoor localization. Technologies and techniques can be employed in combination, aimed to improve accuracy and precision of localization by heterogeneous data fusion. Object recognition, moving object localization and tracking can be successfully implemented using integration of RFID technology and digital image processing techniques. The block matching algorithm based on region of interest can be efficiently used in image processing analysis for motion segmentation and object tracking. By using regions of interest we eliminate the influence of other large moving objects and avoid unnecessary computations. In this chapter, the improvement of RFID localization using motion segmentation applied on the region of interest extracted using RFID is described. The presented solution shows significant reduction of the position estimation error and variance in comparison to the conventional passive RFID position estimation.

Miloš Ljubojević, Zdenka Babić, Vladimir Risojević

A Particle Filter Framework for Object Tracking Using Visual-Saliency Information

Abstract

Automated processing of video streams is core to current surveillance systems. The basic building blocks of video processing techniques are object detection and tracking. Tracking results are further analyzed to detect various events and activities for situation assessment. Several approaches to object detection and tracking are based on background modeling. These approaches are generally vulnerable to noise, illumination changes etc. Further, the object may not look similar in an image sequence over time due to changes in orientation, lighting, occlusion, etc. In this chapter, we explore application of neurobiology-saliency for object detection and tracking using particle filters. We use low-level features such as color, luminance and edge information along with motion cues to track a single person. Experimental results show that this approach is illumination invariant and can track persons in varying lighting conditions.

Dwarikanath Mahapatra, Mukesh Saini

Multiresolution Depth Map Estimation in PTZ Camera Network

Abstract

In this chapter, an active stereo vision system composed of two pan-tilt-zoom (PTZ) cameras is proposed for estimating multiresolution depth map for a large and complex scene. The rectification of stereo images is performed based on the sigmoid interpolation with a set of neural networks. The orientation parameters (pan and tilt values) and the rectification transformations of corresponding images are used as the input-output pairs for network training. The input data is read from cameras directly, whereas the output data is computed offline. The trained neural network is used to interpolate rectification transformations in real time for the stereo images captured at arbitrary pan and tilt settings. The correspondence between the stereo images is obtained using a chain of homographies based scheme. Non-homogeneity between the intrinsic parameters of two cameras is treated by means of zoom compensation to improve the quality of stereo rectification. Experimental results are given for estimating multiresolution depth map for a scene.

Sanjeev Kumar, Christian Micheloni, Balasubramanian Raman

Performance Evaluation in Video-Surveillance Systems: The EventVideo Project Evaluation Protocols

Abstract

During recent years, automatic video-surveillance systems have experienced a great development driven by the growing need for security. Many approaches exist whose performance is not clear for a large variety of available scenarios. To precisely identify which ones operate better for each scenario, empirical performance evaluation has been widely used for determining their strengths and weaknesses through their results. This approach requires defining two aspects (usually named as the evaluation protocol): the dataset (representative sequences) and the metrics (performance estimators). Common empirical approaches use metrics based on ground-truth data that define an ideal result, but there are also some novel approaches that do not require such data. Furthermore, the existence of several metrics and the growing availability of video data increase the complexity of the protocol design as well as require us to automate the whole evaluation process. In this chapter, considering the main analysis stages of a typical video-surveillance system (video object segmentation, people detection, video object tracking and event recognition), we introduce their evaluation protocols within the scope of the EventVideo project.

Juan C. SanMiguel, Álvaro García-Martín, José M. Martínez

Titel: Intelligent Multimedia Surveillance
herausgegeben von: Pradeep K. Atrey
Mohan S. Kankanhalli
Andrea Cavallaro
Verlag: Springer Berlin Heidelberg
Electronic ISBN: 978-3-642-41512-8
Print ISBN: 978-3-642-41511-1
DOI: https://doi.org/10.1007/978-3-642-41512-8