Deepfake and Image Forgery Detection

Name: Deepfake and Image Forgery Detection | Cybersecurity, Multimedia Forensics, Image Manipulation
Brand: De Gruyter
Price: 89.95 EUR
Availability: OnlineOnly

Cybersecurity, Multimedia Forensics, Image Manipulation

Chad Johnson(Author)

De Gruyter (Publisher)

1st Edition

Published on 21. April 2026

XXV, 438 pages

E-Book

PDF with digital watermarking

System requirements

978-3-11-173436-1 (ISBN)

€89.95incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Person

Content

Intro
Acknowledgments
Contents
1 Introduction to Deepfakes and Image Forgeries
1.1 Definition and Scope
1.1.1 A Taxonomy of Images
1.1.1.1 Techniques Versus Intent
1.1.1.2 Camera Original and Other Unaltered Images
1.1.1.3 Types of Altered Images
1.1.1.4 Why the Taxonomy Matters
1.2 Historical Context and Evolution
1.3 Impact on Society and Media
1.3.1 Political Manipulation in Practice
1.3.2 Celebrity and Public Figure Impersonation
1.3.3 Financial Fraud and Security Breaches
1.3.4 Erosion of Public Trust
1.3.5 Regulatory and Policy Response
1.3.6 Economic Costs and Emerging Markets
1.3.7 Legal Evidence and Courtroom Admissibility
1.3.8 Child Safety Implications
1.3.9 Mental Health Impact on Victims
1.3.10 Disinformation in Conflict Zones
1.3.11 Journalistic Verification Workflows
1.3.12 Educational Consequences
1.3.13 Cultural Heritage and Art Market Fraud
1.4 Foundational Signals: What They Are and Why They Matter
1.5 Challenges and the Need for Detection
1.6 How to Use This Book
1.6.1 Why This Book Exists
1.6.2 What This Book Covers-and What It Doesn't
1.6.3 Foundational Ideas You'll Meet Immediately
1.6.4 How to Use This Book
1.6.5 Learning Outcomes
1.6.6 The Map: From Physics to Testimony
1.6.7 Standards and Admissibility
1.6.8 Ethical Stance
1.6.9 Prerequisites and Study Advice
1.6.10 A Final Word on Mindset
1.7 Summary of Terms
Sources
2 Principles of Multimedia Forensics
2.1 Foundational Concepts
2.1.1 Definition and Key Objectives
2.2 Aspects of an Image That Contains Forensic Artifacts
2.2.1 Substantive Qualities
2.2.1.1 Subject and Scene
2.2.1.1.1 Subject-Driven: Figural Images
2.2.1.1.2 Subject-Driven: Selfies
2.2.1.1.3 Subject-Driven: Still Lifes
2.2.1.1.4 Subject-Driven: Portrait
2.2.1.2 Scenic and Subjectless Images
2.2.1.2.1 Scene-Driven: Empty Place Images
2.2.1.2.2 Scene-Driven: Abstract Images
2.2.1.2.3 Scene-Driven: Incidental or Unintended Images
2.2.1.3 Secondary Leakage
2.2.1.3.1 Scale
2.2.1.3.2 Distance
2.2.1.3.3 Temporal
2.2.1.3.3.2 Indirect Cues
2.2.1.3.3.3 External Anchors
2.2.1.3.4 Geographical (Localisms)
2.2.1.3.5 Slippage
2.2.1.3.5.2 Countenance
2.2.1.3.5.3 Reflections
2.2.2 Contextual Qualities
2.2.2.1 Provenance
2.2.2.2 Intentionality
2.2.2.3 Disposition
2.2.2.4 Metadata
2.2.2.4.1 File System Metadata
2.2.2.4.1.1 Practical Forensic Implications (What Examiners Should Watch)
2.2.2.4.2 EXIF Metadata
2.2.2.4.3 Sidecar File Metadata
2.2.2.5 Relationality
2.2.2.5.1 Classifications of Relationality
2.2.2.5.2 Content-Equivalence Labels
2.2.2.5.3 Evidence Tags
2.2.2.5.3.2 Sample Notes and Summary
2.2.2.5.4 Lossy Image Quantization Values
2.2.2.5.5 A Relationality Example
2.3 PRNU (Photo-Response Nonuniformity): A Plain-Language Primer
2.3.1 How Analysts Use PRNU
2.3.2 What PRNU Is-and Isn't
2.3.3 When PRNU Gets Tricky
2.3.4 Can PRNU Itself Be Forged or Removed?
2.3.5 Where PRNU Comes from in the Imaging Chain
2.3.6 How We Explain a PRNU Finding
2.4 JPEG Foundations: DCT Blocks and Quantization
2.5 Recompression and Double Compression
2.6 The Discrete Cosine Transform (DCT)
2.7 Quantization and Quantization Tables
2.8 Chroma Subsampling (4:4:4, 4:2:2, 4:2:0)
2.9 MCUs and the JPEG Grid
2.10 Entropy Coding and Marker Fingerprints
2.11 RGBYCbCr and Transfer Functions (Gamma)
2.12 Resampling and Interpolation Kernels
2.13 Demosaicing and CFA Patterns
2.14 Rolling Shutter and Temporal Sampling
2.15 Perceptual Hashing Versus Cryptographic Hashing
2.16 "ELA" and Transform-Domain Thinking
2.17 JPEG Versus HEIC/AVIF (Modern Codecs)
2.18 Dynamic Range, Bit Depth, and Banding
2.19 Conclusion
2.20 Summary of Terms
Sources
3 Methods of Image Manipulation
3.1 Types of Digital Manipulation
3.2 Pixel-Level Alterations
3.2.1 Brightness, Contrast, and Color Correction
3.2.2 Local Tone Mapping
3.2.3 Gamma Adjustment
3.2.4 Blurring, Sharpening, and Noise Insertion/Removal
3.2.5 Stylization
3.2.6 Local Region Enhancements (e.g., Teeth Whitening and Skin Smoothing)
3.3 Geometric Transformations
3.3.1 Rotation, Scaling, and Skewing
3.3.2 Cropping and Reframing
3.3.3 Flipping or Mirroring
3.3.4 Lens Correction and Intentional Distortion
3.3.5 Resampling Artifacts
3.3.6 A Practical Examination Flow
3.4 Copy-Move Forgery (Clone Stamping)
3.4.1 Patch-Based Duplication
3.4.2 Texture Cloning
3.4.3 Region-Based Detection
3.4.4 Keypoint-Based Block Matching
3.4.5 Block-Based Self-Similarity Detection
3.4.6 Dense Field Detection and Offset Analysis
3.4.7 Region Growing and Transformation Clustering
3.4.8 Clone Inconsistency in Texture, Noise, or Geometry
3.5 Splicing and Compositing
3.5.1 Hard- and Soft-Edge Blending
3.5.2 Shadow and Lighting Inconsistencies
3.5.3 Focus and Depth-of-Field Mismatch
3.5.4 Noise and Compression Inconsistencies
3.6 Traditional Editing and Retouching
3.6.1 Healing Brush Artifacts
3.6.2 Airbrushing and Smoothing Tools
3.6.3 Edge Enhancement Exaggeration
3.6.4 Color, White Balance, and Gamut
3.6.5 Reflections, Occlusions, and Contact Geometry
3.6.6 Resampling and Matting Signatures
3.7 AI-Based and Synthesis Manipulations
3.7.1 AI-Generated and GAN-Based Forgeries
3.7.2 Deepfake Facial Reenactment
3.7.3 Entire Image Generation (StyleGAN, DALL · E, and Midjourney)
3.7.4 Face Morphing, Age Progression, Expression Transfer
3.7.5 Detection by GAN Fingerprinting, Color Statistics, or Reverse Model Attribution
3.8 Anti-forensic Techniques
3.8.1 Anti-forensics
3.8.2 Metadata Removal, Spoofing, or Replacement
3.8.3 Histogram Equalization to Conceal Tampering
3.8.4 Recompression to Erase Copy-Move Footprints
3.8.5 Intentional Noise or Watermark Disruption
3.9 Hybrid Manipulations
3.9.1 Hybrid Techniques
3.9.2 Multi-method Integration
3.9.3 Intentional Overlap of Methods
3.9.4 Layered Compositing, Retouching, and GAN Polishing
3.10 Video-Specific Manipulations
3.10.1 Temporal Tampering
3.10.2 Inter-frame Forgery
3.10.3 Intra-frame Manipulation
3.10.4 Speed and Duration Manipulation
3.10.5 Audio Manipulation and Mismatch
3.10.6 Compression Artifacts as Indicators
3.11 Summary of Terms
Sources
4 Best Practices and Forensic Procedure
4.1 A Standard-Driven Foundation
4.1.1 Admissibility and Gatekeeping (Daubert, FRE 702, and Beyond)
4.1.2 Validation Artifacts (Produce These on Every Case)
4.1.3 Reporting Artifacts and Document Types
4.1.3.1 Report Types Your Lab Should Standardize
4.2 Evidence Preservation and Chain of Custody in Digital Media
4.2.1 Authenticity and Integrity
4.2.2 Chain of Custody
4.2.3 Validity Without Matching Hashes
4.2.4 Secure Storage and Non-repudiation
4.2.4.1 Establishing and Preserving Chain of Custody
4.2.4.2 Forensic Acquisition and Environment Control
4.2.4.3 Tool Validation and Quality Assurance
4.2.4.4 Structured Examination and Analysis
4.2.4.5 Interpretation, Peer Review, and Technical Reporting
4.2.4.6 Presentation and Testimony
4.2.4.7 Archiving and Long-Term Preservation
4.2.5 Closing Observations
4.3 Scientific Methodologies
4.3.1 Analytical Frameworks in Forensic Investigations
4.3.2 Comparison Between Traditional and Digital Forensics
4.3.2.1 Intake and Preservation
4.3.2.2 Survey
4.3.2.3 Blind Recall Memo
4.3.2.4 Scope and Examination Plan
4.3.2.5 Substantive Analysis (Close Examination)
4.3.2.6 Contextual and Provenance Analysis
4.3.2.7 Container and Compression Forensics
4.3.2.8 Statistical and Feature-Based Tests
4.3.2.9 Sensor and Optical Signature Analysis
4.3.2.10 Temporal and Geospatial Reconstruction
4.3.2.11 Cross-Asset Corroboration
4.3.2.12 Falsification/Red-Team Pass
4.3.2.13 Synthesis and Confidence Statement
4.3.2.14 Reporting and Exhibits
4.3.2.15 Peer/Legal Review and Readiness
4.3.2.16 Archival and Disclosure
4.4 Structured Examination
4.5 Interpretation, Reporting, and Testimony
4.6 Reporting Language and Legal Gatekeeping
4.7 Archiving
4.8 Bias, Standards, and Provenance in Practice
4.9 Tools and Technologies
4.9.1 Blind Proficiency as Routine, Not Event
4.9.2 Display Fidelity and Color Management
4.10 Summary of Terms
Sources
5 Methods of Image Authentication
5.1 Active Authentication: Preventing Ambiguity Before It Begins
5.1.1 Substantive and Contextual Analysis
5.1.2 Feature-Based and Statistical Analysis
5.1.3 Compression and Encoding Forensics
5.1.4 Sensor and Optical Signature Analysis
5.1.5 Motion and Temporal Analysis
5.2 Integrating the Methods
5.3 Methods of Image Authentication
5.4 Overview of Authentication Techniques
5.4.1 Digital Signatures
5.4.2 Watermarks
5.4.3 Cryptographic Hash Functions
5.4.4 Blockchain-Based Provenance Systems
5.5 Comparison of Traditional and Modern Approaches
5.6 Algorithmic Approaches
5.6.1 Machine Learning, Statistical Models, and Analytical Detection
5.6.2 Machine Learning Models
5.6.3 Statistical Inference and Signal Analysis
5.6.4 Error Level Analysis and Visual Consistency Testing
5.7 Challenges and Future Prospects
5.7.1 Limitations of Contemporary Techniques and the Road Ahead
5.7.2 Vulnerability to Sophisticated Attacks
5.7.3 False Positives and Negatives
5.7.4 Scale, Speed, and Resource Constraints
5.8 Directions for Enhancing Robustness
5.8.1 Architectures, Adaptation, and the Next Generation of Tools
5.8.2 Hybrid Systems
5.8.3 Adaptive Algorithms and Continuous Learning
5.8.4 Emerging Technologies and Infrastructure
5.8.5 Standardization and Collaborative Development
5.9 Conclusion
5.10 Summary of Terms
6 Substantive and Contextual Analysis
6.1 Surface-Level Visual Anomalies: Perceptual Coherence
6.1.1 Lighting and Shadow Inconsistencies
6.1.2 Color Temperature and White Balance
6.1.3 Depth of Field and Focus Mismatch
6.1.4 Compression Artifacts and Resolution Mismatch
6.1.5 Noise and Grain Inconsistencies
6.2 Spatial Integration and Physical Plausibility
6.2.1 Perspective and Spatial Relationships
6.3 Rooting and Contact Physics
6.3.1 Seaming and Webbing
6.3.2 Environmental Inconsistencies
6.3.3 Common Cues and Contradictions
6.3.4 Analytical Techniques
6.4 Biological and Anatomical Cues
6.4.1 Anatomical and Biological Inconsistencies
6.4.2 Typical Biological Errors
6.4.3 Techniques and Tools
6.4.4 Specular Highlights and Reflections
6.4.4.1 Key Indicators of Inconsistency
6.4.4.2 Detection and Analysis Techniques
6.4.5 Emotion and Expression Cues
6.4.5.1 Indicators of Emotional or Expression-Based Inconsistencies
6.4.5.2 Forensic Interpretation
6.5 Symbolic and Cultural Plausibility
6.5.1 Fashion, Technology, and Cultural Anachronisms
6.5.2 Common Anachronisms and Cultural Incongruities
6.5.2.1 Forensic Approaches
6.5.3 Text and Symbol Errors
6.5.4 Contextual Plausibility and Narrative Coherence
6.6 Metadata, Provenance, and Contradictions
6.6.1 Metadata-Contradictory Content
6.7 Summary of Terms
Sources
7 Feature-Based Analysis: Extraction and Identification
7.1 Hashing and Robust Feature Matching
7.1.1 Cryptographic Hashes (Fragile)
7.1.2 Perceptual Hashing (pHash, aHash, and dHash)
7.1.3 Robust Local Features (SIFT, SURF, and ORB)
7.1.4 Global Descriptors and Learned Embeddings
7.1.5 Applications in Forensics
7.1.6 Limitations and Best Practices
7.2 Frequency-Domain Analysis
7.2.1 Frequency-Domain Analysis: Seeing Beyond the Pixels
7.3 Audio Consistency Checks
7.4 Texture Analysis
7.5 Texture Analysis for Image Forgery Detection
7.5.1 What Is Texture in a Forensic Context?
7.6 Texture Coherence and Internal Consistency
7.6.1 Common Texture Anomalies in Forged Media
7.6.2 Tools and Techniques for Texture Analysis
7.6.3 Integrating Texture Analysis with Other Forensic Methods
7.7 Statistical Methods for Texture Characterization
7.7.1 Tools
7.8 Copy-Move Detection and Self-Similarity in Texture
7.9 Keypoint Descriptor Comparison in Image Forgery Detection
7.9.1 What Are Keypoints and Descriptors?
7.9.2 SIFT (Scale-Invariant Feature Transform)
7.9.3 SURF (Speeded-Up Robust Features)
7.9.4 ORB (Oriented FAST and Rotated BRIEF)
7.9.5 Descriptor Matching Techniques
7.9.6 Geometric Consistency Verification
7.9.7 Visualizing and Validating Matches
7.10 Edge and Contour Consistency in Forgery Detection
7.10.1 Understanding Edge Semantics
7.10.2 Forensic Cues Derived from Edge Analysis
7.10.3 Algorithmic Edge Analysis
7.10.4 Application in Copy-Move and Splicing Detection
7.10.5 Visual Acuity and Training
7.11 Deep Learning and Learned Texture Models
7.11.1 Texture Representation in Deep Neural Networks
7.11.2 Detecting GAN-Generated Texture Anomalies
7.11.3 GAN Fingerprinting with Texture Signatures
7.11.4 Applications and Integration
7.12 Texture-Based Segmentation and Localization
7.12.1 Understanding Texture-Based Segmentation
7.12.2 Techniques for Localization
7.12.3 Visualization and Human Interpretation
7.12.4 Integration with Other Modalities
7.13 Texture Transfer and Stylization Artifacts
7.14 Manual Texture Inspection Techniques
7.15 Summary of Terms
Sources
8 Compression and Encoding Forensics
8.1 DCT Coefficient Mapping and Quantization Signatures
8.2 JPEG Artifacts and Noise Analysis
8.3 Understanding JPEG Compression
8.3.1 Double JPEG Compression and Grid Misalignment
8.4 Noise Profiles and Sensor Fingerprints
8.5 Differentiating Real from GAN-Based Images
8.6 Practical Considerations and Caveats
8.7 Digital Noise: Structure, Sources, and Forensic Value
8.7.1 Types of Noise and Their Origins
8.7.2 Camera-Specific Noise Patterns
8.7.3 Photographic Practices That Modify Noise
8.7.4 Noise Discontinuities and Tampering Detection
8.7.5 Limitations and Considerations
8.8 Error Level Analysis (ELA)
8.9 Double-Compression Detection
8.10 Double JPEG Compression Artifacts
8.10.1 How JPEG Compression Works
8.10.2 What Happens in Double Compression
8.10.3 Forensic Detection Techniques
8.10.4 Practical Considerations and Limitations
8.10.5 Applications
8.11 Camera Signature-Based Detection
8.11.1 Physical Signatures: Sensor Pattern Noise (SPN) and PRNU
8.11.2 Processing Signatures: CFA, Demosaicing, and Compression Artifacts
8.11.3 Source Camera Attribution and Anti-forensics
8.12 CFA Pattern Analysis and Demosaicing Consistency
8.12.1 The Bayer Pattern and Color Subsampling
8.12.2 Detection Techniques and Analytical Tools
8.12.3 Common Forensic Scenarios
8.12.4 Practical Considerations and Limitations
8.13 Chromatic Aberration Analysis
8.13.1 Forensic Use of Chromatic Aberration
8.14 Lens Distortion Patterns
8.14.1 Forensic Application
8.14.2 Integration with Other Analyses
8.15 JPEG Ghost Detection
8.15.1 Principle of JPEG Ghost Detection
8.15.2 Operational Methodology
8.15.3 Analytical Interpretation
8.15.4 Technical Considerations
8.15.5 Applications
8.15.6 Tool Support
8.16 Quantization Table Analysis
8.16.1 Understanding Quantization Tables in JPEG Compression
8.16.2 Quantization Table as Forensic Evidence
8.16.3 Applications in Forgery Detection
8.17 Quantization Table Fingerprinting
8.17.1 Quantization Table Tampering and Anti-forensics
8.17.2 Analytical Techniques
8.17.3 Limitations and Considerations
8.18 Conclusion
8.19 Summary of Terms
Sources
9 Sensor and Optical Signature Analysis
9.1 Sensor Fingerprinting: PRNU Without the Mystery
9.1.1 Forensic Value
9.1.2 Getting the Fingerprint (Step by Step)
9.1.2.1 Practice Notes
9.1.2.2 Practical Tips
9.1.3 What PRNU Is-and Where It Comes from
9.1.4 Matching a Questioned Image
9.1.5 Localizing Tampering
9.1.5.1 Source Match
9.1.5.2 Alignment Search
9.1.5.3 Tampering Maps
9.1.5.4 Expanded Caution
9.1.6 Failure Modes, Confounders, and Modern Pipelines
9.1.7 Security Considerations: Copy Attacks and Defenses
9.1.8 Reporting and Courtroom Framing
9.2 CFA and Demosaicing Signatures
9.2.1 Why CFA Traces Exist (and Why They Survive)
9.2.2 What to Look for (Observable Signatures)
9.2.3 Core Analytics (How to Quantify)
9.2.4 A Lab-Ready Workflow (with Thresholds and Exhibits)
9.2.5 Handling Real-World Complications
9.2.6 Integrating with Resampling and Copy-Move Detectors
9.2.7 Validation, Thresholds, and Reporting
9.2.8 Quick-Reference Checklist (CFA/Demosaicing)
9.3 Optical System Signatures: Vignetting, Distortion, Chromatic Aberration, PSF/Bokeh
9.3.1 Why Optical Traces Matter
9.3.2 Vignetting (Radial Falloff)
9.3.3 Geometric Distortion (Barrel, Pincushion, and Moustache)
9.3.4 Chromatic Aberration (CA) Maps
9.3.5 Spatially Varying PSF and Bokeh
9.3.6 A Unified Optical-Forensic Workflow
9.3.7 Real-World Complications (and How to Handle Them)
9.3.8 Integration with PRNU and CFA
9.3.9 Validation and Thresholds
9.3.10 Quick Checklist (Optics)
9.4 Readout and Shutter Artifacts: Rolling Shutter, Row/Column Noise, Mains Flicker
9.4.1 Why Readout Artifacts Matter
9.4.2 Rolling Shutter: Geometry and Observables
9.4.3 Estimating Rolling-Shutter Parameters in Practice
9.4.4 Using Rolling-Shutter Coherence to Find Splices
9.4.5 Mains-Rate Flicker: Phase and Stripe Analysis
9.4.6 Row/Column Fixed-Pattern Noise (FPN) and Banding
9.4.7 Practical Workflow (Video and High-Frame-Rate Stills)
9.4.8 What Adversaries Do-and Counters
9.4.9 Validation and Reporting
9.5 Computational Photography: When Phones Rewrite the Signal
9.5.1 Why This Matters for Forensics
9.5.2 What Contemporary Pipelines Actually Do to the Signal
9.5.3 Recognizing Computational Processing in Practice
9.5.4 Adapting Attribution and Tamper Localization
9.5.5 Distinguishing ISP Corrections from Human Edits
9.5.6 Super-Resolution and Learned Hallucination as a Confound and a Clue
9.5.7 Night-Mode Specifics
9.5.8 Anti-forensics That Exploit These Pipelines-and Counters
9.5.9 Validation, Thresholds, and Report Language
9.6 PRNU for Video
9.7 Putting It Together: A Practical Recipe
9.8 How to Talk About This Under Daubert/FRE 702
9.9 Color Filter Array (CFA) and Demosaicing Signatures
9.9.1 What Exactly Survives the Pipeline
9.9.2 How to Measure It Defensibly
9.9.3 What Splices and Resampling Do to These Cues
9.9.4 Non-Bayer Layouts and Modern Phone Pipelines
9.9.5 Failure Modes and How to Avoid Them
9.9.6 A Compact Workflow You Can Defend
9.10 Optical System Signatures: Vignetting, Distortion, CA, PSF
9.11 Readout and Shutter Artifacts: Rolling Shutter, Row/Column Banding, Flicker, Defect Constellations
9.11.1 Rolling-Shutter Geometry You Can Measure
9.11.2 Row/Column Pattern Noise and Fixed-Frequency Banding
9.11.3 Illumination Flicker and Row-Time Aliasing (Video)
9.11.4 A Defensible Workflow
9.11.5 Sensor Defect Constellations
9.11.6 Limits and Integration
9.12 Illumination and Flash Signatures Tied to the Optics and the Sensor
9.12.1 Radial Falloff Under On-Camera Lighting
9.12.2 Catchlights and Specular Geometry in Eyes and Mirrors
9.12.3 Spectral Fingerprints: IR-Cut, Color Pipeline, and White Balance
9.12.4 Bringing the Cues Together in Analysis
9.12.5 Confounders and Limits
9.13 Computational Photography Effects (and How to Handle Them)
9.13.1 What These Pipelines Do to Sensor and Optical Traces
9.13.2 Recognizing Computational Footprints in Practice
9.13.3 Handling Strategies: Adapting Tests and Reporting Limits
9.13.4 A Short Decision Pattern for Examiners
9.14 Video PRNU and Cross-Modal Linkage
9.15 Anti-forensics, Validation, and Calibrated Conclusions
9.16 Standard Operating Procedure: Building and Using a PRNU System
9.16.1 Scope and Preconditions
9.16.2 Reference Construction (Device Fingerprint)
9.16.3 Operating Point and Sanity Panel
9.16.4 Case Ingestion and Global Attribution
9.16.5 Localization (Tamper Mapping)
9.16.6 Anti-forensic Screening
9.16.7 Corroboration Outside the Sensor Domain
9.16.8 Quality Assurance, Reproducibility, and Change Control
9.16.9 Reporting and One-Page Validation Sheet
9.17 Summary of Terms
Sources
10 Motion and Temporal Analysis
10.1 Optical Flow Analysis
10.1.1 Theoretical Foundations
10.1.2 Core Algorithms
10.2 Forensic Applications of Optical Flow
10.2.1 Visualizing and Interpreting Flow
10.2.2 Method Limits and the Error Budget
10.2.3 Tools That Hold Up in Practice
10.3 Temporal Inconsistencies
10.3.1 Replay Attacks (Screen-on-Camera)
10.3.2 Two- Versus Three-Dimensional Face Cues
10.3.3 Physiological Signals (rPPG)
10.3.4 Challenge-Response (When You Control Capture)
10.3.5 Deepfake Video Liveness
10.4 Liveness Detection
10.4.1 Container and Bitstream Timeline
10.4.2 Frame-Level Continuity
10.4.3 Scene-Level Time Cues
10.4.4 Cross-Modal Timing (A/V Sync)
10.5 Court-Ready Workflow (Motion and Temporal)
10.5.1 Motion Inconsistency Detection
10.5.2 Scene-Flow Discontinuities (Global vs. Local Motion)
10.5.3 Temporal Tampering (Frame Insertion/Deletion/Retiming)
10.5.4 Deepfake and Facial Reenactment Cues
10.5.5 Stabilization Artifacts and Recompression Fingerprints
Sources
11 Learned Detection and Hybrid Systems
11.1 Deepfake Detection Models
11.2 Robust Features
11.3 Sparse Representation
11.4 Dimensionality Reduction
11.5 Dynamic Thresholding
11.6 Putting It Together: Hybrid Systems You Can Defend
11.7 Adversarial and Anti-forensic Pressure (and Defenses)
11.8 Provenance and Watermark Signals (Complements, Not Substitutes)
11.9 Benchmarking, Generalization, and Reporting for Learned Systems
11.10 Summary of Terms
Sources
12_Johnson_0426_EF_Ch12
12 Legal and Ethical Considerations
12.1 Expression, Privacy, and the Democratic Commons
12.1.1 Access, ToS, and Anti-circumvention
12.2 Torts, Remedies, and Civil Exposure
12.2.1 Spoliation and Legal Holds
12.3 Evidence Law and Admissibility
12.3.1 Rule 702 (2023 Amendments), Frye, and Kumho
12.3.2 Self-Authenticating Electronic Records and Summaries
12.4 Provenance Tech, Platforms, and the "Liar's Dividend"
12.5 Children, CSAI, and Mandatory Reporting
12.6 Professional Duty: Rigor with Restraint
12.6.1 Testifying Versus Consulting Expert
12.6.2 Language Discipline
12.7 Data Protection, Biometrics, and Consent
12.7.1 Cross-Border Transfers
12.8 Research Ethics, Datasets, and Dual Use
12.9 Work with Journalists and Civil Society
12.10 Operational Security and Practitioner Safety
12.11 Putting It Together
12.12 Summary of Terms
Sources
13_Johnson_0426_EF_Ch13
13 Future Directions in Image Forgery Detection
13.1 Diffusion-Native Detection and Open-World Robustness
13.2 Cryptographically Anchored Provenance
13.3 Watermarking That Survives the Wild
13.4 Neuromorphic and Secure-Capture Hardware
13.5 Federated and Privacy-Preserving Learning
13.6 Real-Time, Edge-Native Verification
13.7 Holistic, Multimodal Forensics
13.8 Explainability, Calibration, and Legal Readiness
13.9 A Convergent Outlook
13.10 Summary of Terms
Sources
14 Glossary
Sources
Appendix A: Case Report Template
Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Deepfake and Image Forgery Detection

Description

More details

Other editions

Additional editions

Person

Content

System requirements