Recent Papers

With the rapid evolution of digital imaging and editing technologies, image forgery has emerged as a critical threat to information integrity, posing serious challenges in fields such as forensics, journalism, and authentication systems. Traditional detection techniques often fall short when faced with sophisticated manipulation methods like copy-move, splicing, and deepfakes. To overcome these limitations, this study introduces a novel image forgery detection framework that leverages the fusion of multiple lightweight deep learning models.Unlike conventional approaches that rely on complex and resource-intensive architectures, the proposed method integrates several efficient neural networks to enhance detection accuracy while maintaining computational efficiency. The framework employs a multi-branch feature extraction module, where each branch utilizes a distinct lightweight model to capture complementary spatial and frequency domain features. These diverse features are then fused through an attention-based mechanism that highlights forgery-affected regions, enabling precise localization and classification of tampered content.The effectiveness of the proposed approach is validated through extensive experiments on benchmark datasets, including CASIA, CoMoFoD, and DEFACTO. Results demonstrate that our method consistently outperforms existing state-of-the-art techniques in terms of accuracy, precision, recall, and F1-score, all while keeping the computational cost lowmaking it well-suited for real-time applications and deployment in resource-constrained environments.Additionally, ablation studies are conducted to evaluate the individual contributions of each model component and fusion strategy, offering valuable insights into the systems design. This research contributes a robust, scalable, and efficient solution for image forgery detection, advancing the capabilities of modern image forensics across a wide range of practical scenarios.Keywords Image forgery detection, deep learning, lightweight neural networks, feature fusion, attention mechanism, copy-move forgery, splicing detection, deepfake detection, image forensics, computational efficiency.

Copyright © 2025 Mrs. K. Sandhya Rani. This is an open access article distributed under the Creative Commons Attribution License.