دانلود کتاب Digital Forensics and Watermarking: 21st International Workshop, IWDW 2022, Guilin, China, November 18-19, 2022, Revised Selected Papers (Lecture Notes in Computer Science, 13825)

فهرست مطالب :

Preface\nOrganization\nContents\nSteganology\nHigh-Performance Steganographic Coding Based on Sub-Polarized Channel\n 1 Introduction\n 2 Preliminaries\n 2.1 The Theory of Optimal Steganographic Embedding\n 2.2 Channel Polarization\n 2.3 Decoding of Polar Codes\n 3 Steganographic Coding on Sub-Polarized Channel\n 3.1 Polarized Steganographic Channel\n 3.2 Successive Cancellation on Polarized Steganographic Channel\n 3.3 Steganographic Coding Under the Typical Sub-Channel\n 4 Experimental Results\n 4.1 Construction of Polarized Steganographic Channel\n 4.2 Security Evaluation Under Embedding Efficiency Results\n 4.3 Security Evaluation Under Image Steganalysis\n 4.4 Evaluation of the Computational Efficiency\n 4.5 Discussion\n 5 Conclusion\n References\nHigh-Capacity Adaptive Steganography Based on Transform Coefficient for HEVC\n 1 Introduction\n 2 Analysis of Embedding Distortion\n 2.1 Embedding Distortion of Visual Degradation\n 2.2 Embedding Distortion of GOP\n 3 Proposed HEVC Steganography\n 3.1 Selection of Cover Coefficients\n 3.2 Proposed Cost Function\n 3.3 The Practical Implementation\n 4 Experimental Results\n 4.1 Experimental Setup\n 4.2 Performance of Visual Quality\n 4.3 Performance of Embedding Capacity\n 4.4 Performance of Anti-steganalysis\n 5 Conclusion\n References\nForensics and Security Analysis\nSE-ResNet56: Robust Network Model for Deepfake Detection\n 1 Introduction\n 2 Related Works\n 2.1 Physiological Features Based Deepfake Detection Method\n 2.2 Specific Artifact Based Deepfake Detection Method\n 2.3 Data-Driven Deepfake Detection Method\n 3 The Proposed Method\n 3.1 Improved Residual Block SE-Res-Block\n 3.2 Network Framework\n 3.3 Deepfake Face Image Detection Algorithm Based on SE-RESNET56\n 4 Experimental Results and Performance Analysis\n 4.1 Evaluation Indicators and Dataset Settings\n 4.2 Ablation Experiment\n 4.3 The Performance Analysis\n 4.4 Robustness Analysis\n 4.5 Generalization Analysis\n 5 Conclusion\n References\nVoice Conversion Using Learnable Similarity-Guided Masked Autoencoder\n 1 Introduction\n 2 Related Work\n 3 Backgroud\n 3.1 Feature Decoupling for SII\n 4 Method\n 4.1 LSGM\n 4.2 MAE-VC\n 5 Experiments and Results\n 5.1 Experiment Conditions\n 5.2 Metrics\n 5.3 Results and Discussions\n 6 Conclusion\n References\nVisual Explanations for Exposing Potential Inconsistency of Deepfakes\n 1 Introduction\n 2 Related Works\n 2.1 Deepfake Detection\n 2.2 Vision Transformer\n 3 Method\n 3.1 Feature Enhancement\n 3.2 Multi-scale Feature Extraction\n 3.3 Classification\n 3.4 SPI\n 4 Experiments\n 4.1 Experiment Setting\n 4.2 Comparison Results\n 4.3 Distinguish Different Forgeries\n 4.4 Robustness to Video Compression\n 4.5 Visualization Analysis\n 5 Conclusions\n References\nImproving the Transferability of Adversarial Attacks Through Both Front and Rear Vector Method\n 1 Introduction\n 2 Related Work\n 2.1 Adversarial Attack Methods\n 2.2 Adversarial Training\n 3 Methodology\n 3.1 Previous-Gradient and Momentum as Neighborhood NI-FGSM\n 3.2 Both Front and Rear Vector Method\n 4 Experiments\n 4.1 Setup\n 4.2 Attack Single Models\n 4.3 Attack Ensemble Models\n 4.4 Ablation Study\n 5 Conclusion\n References\nManipulated Face Detection and Localization Based on Semantic Segmentation\n 1 Introduction\n 2 Related Works\n 2.1 Face Manipulation\n 2.2 Manipulated Face Detection\n 2.3 Manipulated Face Localization\n 3 Methodology\n 3.1 Labeled Data Generation\n 3.2 Multi-branch Autoencoder\n 3.3 Loss Function\n 4 Experiments\n 4.1 Experimental Setup\n 4.2 Intra-dataset Evaluation\n 4.3 Inter-dataset Evaluation\n 5 Conclusion\n References\nDeep Learning Image Age Approximation - What is More Relevant: Image Content or Age Information?\n 1 Introduction\n 2 Steganalysis Residual Network (SRNet)\n 3 Dataset and Training Settings\n 3.1 PLUS Aging Dataset\n 3.2 Northumbria Temporal Image Forensics Database\n 4 Explainable Artificial Intelligence (XAI)\n 4.1 GradCAM++\n 4.2 ScoreCAM\n 5 CAM Analysis\n 5.1 Activation on Objects\n 5.2 Activation on Areas\n 5.3 No Constant Activation Pattern\n 6 Potential Solutions\n 7 Conclusion\n References\nWatermarking\nPhysical Anti-copying Semi-robust Random Watermarking for QR Code\n 1 Introduction\n 2 Related Work\n 2.1 Physical Anti-copying\n 2.2 Watermarking for QR Code\n 3 Proposed Physical Anti-copying Watermarking System\n 3.1 Model for Authentic and Counterfeit Channels\n 3.2 Watermark Embedding\n 3.3 Watermark Extraction and Authentication\n 4 Experimental Result\n 4.1 Experimental Setup\n 4.2 Comparison of the Authentic and Counterfeited QR Code\n 4.3 Printing Size v.s. Erroneous Bits\n 4.4 Printing Size v.s. Anti-copying Capability\n 4.5 Comparison of Authentication Performance\n 5 Conclusion\n References\nRobust and Imperceptible Watermarking Scheme for GWAS Data Traceability\n 1 Introduction\n 2 Genomic Data and Database Model\n 2.1 Genomic Data\n 2.2 Weighted Sum Statistic (WSS) Method\n 3 Proposed Database Watermarking Scheme for GWAS Data\n 3.1 Database Watermarking\n 3.2 Quantization Index Modulation (QIM)\n 3.3 Watermark Embedding in WSS Data\n 3.4 Watermark Extraction\n 4 Theoretical Performance\n 4.1 Parameter Constraints\n 4.2 Distortion Performance\n 4.3 Robustness Performance\n 5 Experimental Results and Discussion\n 5.1 Test Database\n 5.2 Distortion Results\n 5.3 Capacity Results\n 5.4 Robustness Results\n 6 Conclusion\n References\nAdaptive Robust Watermarking Method Based on Deep Neural Networks\n 1 Introduction\n 2 Proposed Method\n 2.1 Watermark Embedding Module\n 2.2 Watermark Decoding Module\n 2.3 Loss Function\n 3 Experiments\n 3.1 Training Process\n 3.2 Test Results\n 4 Conclusion\n References\nAdaptive Despread Spectrum-Based Image Watermarking for Fast Product Tracking\n 1 Introduction\n 2 Motivation\n 3 Adaptive Despread Spectrum-Based Image Watermarking\n 3.1 Watermark Embedding\n 3.2 Watermark Extraction\n 4 Experimental Evaluation\n 4.1 Experimental Setup\n 4.2 Robustness to Signal Processing Attacks\n 4.3 Robustness to Geometric Attacks\n 4.4 Computational Cost\n 5 Conclusion\n References\nReversible Data Hiding via Arranging Blocks of Bit-Planes in Encrypted Images\n 1 Introduction\n 2 Proposed Method\n 2.1 Vacating Room\n 2.2 Image Encryption\n 2.3 Data Embedding\n 2.4 Data Extraction and Image Recovery\n 3 Experimental Results\n 3.1 Security Analysis\n 3.2 Embedding Performance\n 3.3 Performance Comparison\n 4 Conclusions\n References\nHigh Capacity Reversible Data Hiding for Encrypted 3D Mesh Models Based on Topology\n 1 Introduction\n 2 Related Works\n 3 Proposed Method\n 3.1 Room Reservation\n 3.2 Model Encryption\n 3.3 Data Embedding\n 3.4 Data Extraction and Model Recovery\n 4 Experiment Results and Analysis\n 4.1 Visual Quality and Quantitative Analysis\n 4.2 Capacity Analysis\n 4.3 Features Analysis\n 5 Conclusion\n References\nAuthor Index