Integrating Content Authentication Support in Media Services

Integrating Content Authentication Support in Media Services

Anastasia N. Katsaounidou (Aristotle University of Thessaloniki, Greece) and Charalampos A. Dimoulas (Aristotle University of Thessaloniki, Greece)
Copyright: © 2018 |Pages: 12
DOI: 10.4018/978-1-5225-2255-3.ch254
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The present chapter investigates content authentication strategies and their use in media practice. Remarkable research progress has been conducted on media veracity methods and algorithms, however, without providing that much straightforward tools to users involved in real-world applications. Hence, there is an urgent need for further supporting content verification by exploiting all the available methods in properly integrated online environments, forming a Media Authentication Network. On-demand training (and feedback) on these technologies is considered of major importance, enabling users to collaborate with media and forgery experts towards adoption, refinement and widespread dissemination of best practices. Better comprehension of the involved tools and algorithms would propel their broad exploitation in practice, gaining valuable feedback for further improvements. Thus, a continuously updated online repository, containing documented examples, learning resources and media veracity tools, could be adaptively accommodated, better supporting various users and applications needs.
Chapter Preview
Top

Introduction

The tremendous evolution of Information and Communication Technologies (ICT) and the low cost of the digital media devices have fueled the widespread expansion of the so-called User Generated Content (UGC). Social networking has become a popular way for users to meet and interact online through text and audiovisual content (photos, sounds, videos, etc.) that is produced and distributed in real time. Traditionally, users share news and multimodal content through social media, while simultaneously discover information on the Web. Among others, trust is considered to be one of the crucial factors of information capturing and dissemination. In the road from Web 2.0 to Web 3.0 and beyond, the quality and credibility of the recorded, shared and broadcasted content is controversial (Ljung & Wahlforss, 2008; Matsiola, Dimoulas, Kalliris & Veglis, 2015). Many (easy to use) multimedia capturing and processing tools (desktop applications and online /cloud services) are currently available and can be exploited literately at any time and place through mobile devices. This “processing at the fingertip” vision familiarizes average users with multimodal media production, processing and management tasks (Dimoulas, Veglis & Kalliris, 2014, 2015).

The domination of digital content over traditional analog media (i.e. films and tapes) has given rise to a number of new information security challenges. Digital content can be processed, intentionally altered or falsified and redistributed relatively easy. This has important consequences for governmental, commercial, social and professional media organizations that rely on digital information (Stamm, Wu & Liu, 2013). Hence, mass communication and journalistic processes can be associated to unwanted content tampering, construction of fake evidences, sharing and propagation of untrue stories. In particular, the universality of “digital news reporting” has turned the evaluation of shared media into a field of prime importance, focusing on the automatic detection of manipulated and misused Web content (Mendoza, Poblete & Castillo, 2010). Its aim is to lay the basis for a future generation of tools that could assist media professionals in the process of verification. According to Figure 1, where the problem definition of the discussed topic is presented, media alteration involves all content types (text, images, audio, video, etc.) that are encountered in today's Multimodal Media Assets (Dimoulas et al., 2014; Katsaounidou, 2016).

Figure 1.

Problem definition: (multi)-media tampering as part of digital informing /infotainment

Content alteration can be conducted by anyone involved in the media production processes (media professionals, UGC–citizen journalist, etc.), as the bluish arrows in Figure 1 indicate. Once information falsification occurs without being noticed by the users, uncontrolled propagation of untrue stories may appear as a side effect of the contemporary need for timely and immediate informing. Hence, tampered information can be massively shared /propagated by end-users/consumers (greenish-dotted arrows in Figure 1). The outmost target of the current chapter is to describe a collaborating model for overall supporting content authentication through dedicated computerized environments. In this context, users’ and journalists’ training (and their valuable feedback) holds a key role towards the integration and unification of applicable media veracity services (and their associated learning resources). Thus, algorithms, methodologies and related ground-truth data-sets would be continuously updated, progressed and adapted to the specific needs of the encountered application scenarios.

Key Terms in this Chapter

Digital forensics: A branch of forensic science encompassing the recovery and investigation of material found in digital devices, often in relation to computer crime.

Computer Forensics Science: A research field focusing on content examination with the aim of identifying, preserving and recovering the truth of digital media.

User Generated Content (UGC): Any form of content created and shared by users of an online system or service.

Anti-Forensics: Anti-forensic techniques are content actions aiming at preventing or hardening proper forensic investigation.

Biometrics: Biometric systems measure and analyze various persons’ psycho-physiological parameters (fingerprint, face-expressions, stress-factors, etc.) aiming at evaluating their behavioral status in various cognitive task and/or surveillance scenarios.

Natural Language Processing (NLP): scientific discipline utilizing Machine Learning and Computational Linguistics, aiming at giving computers/machines the ability to perceptually interact trough human (natural) languages.

Machine Learning (ML): Scientific discipline that investigates algorithms and methods aiming at giving machines the ability to learn from experience (without being explicitly programed), in order to respond autonomously on specific tasks and automate various data-handling processes.

Multimodal Media Assets (MMA): The kind of multimedia content commonly encountered in today’s posts and shares, incorporating mixed time-based and page-based media (text, images, audio, video, etc.) along with their metadata.

Digital Image Forensics (DIF): This field emerged as a sub-field of Digital Image Processing (DIP), aiming at providing tools for images tampering investigation.

Complete Chapter List

Search this Book:
Reset