对比四种模型检测假图，VGG16表现最佳
Comparative Evaluation of Deep Learning Models for Fake Image Detection
- 统一预处理与训练流程，比较VGG16等四类模型
- VGG16准确率达91%，其余均达90%
- 提醒需平衡数据集，避免误判真实图像
生成对抗网络（GAN）驱动的图像篡改日益复杂，给数字取证带来挑战。本研究采用统一预处理和训练流程，对比VGG16、ResNet50、EfficientNetB0和XceptionNet四种预训练CNN架构在假图检测中的表现。使用真实与篡改图像数据集，通过调整尺寸、归一化及增强来缓解类别不平衡并提升泛化能力。评估指标包括准确率、精确率、召回率、F1分数和ROC-AUC。结果显示，VGG16准确率最高，达91%；XceptionNet、ResNet50和EfficientNetB0均达到90%。EfficientNetB0对假图更敏感，但在真实样本上可靠性下降，反映出数据不平衡导致的偏差。局限性包括数据分布不均、过拟合及可解释性差，影响跨域鲁棒性。研究提供可复现基线，强调需采用均衡数据集、先进增强策略及公平性感知训练以构建可靠假图检测系统。
原文摘要 · Abstract (English)
The growing sophistication of GAN-based image manipulation presents significant challenges for digital forensics. This study compares the performance of four pretrained CNN architectures including VGG16, ResNet50, EfficientNetB0, and XceptionNet for fake image detection using a unified preprocessing and training pipeline. A dataset of real and manipulated images was processed through resizing, normalization, and augmentation to address class imbalance and improve generalization. Models were evaluated using Accuracy, Precision, Recall, F1-score, and ROC-AUC. VGG16 achieved the highest accuracy at 91%, with XceptionNet, ResNet50, and EfficientNetB0 each reaching 90%. EfficientNetB0 showed stronger sensitivity to fake images but reduced reliability on real samples, reflecting imbalance-driven bias. Limitations include dataset imbalance, overfitting, and limited interpretability, which affect cross-domain robustness. The study provides a reproducible baseline and underscores the need for balanced datasets, advanced augmentation, and fairness-aware training to develop reliable fake image detection systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。