Beyond Pixels: Exploring Deep Learning Methods for Image Forgery Detection
| International Journal of Electronics and Communication Engineering |
| © 2024 by SSRG - IJECE Journal |
| Volume 11 Issue 8 |
| Year of Publication : 2024 |
| Authors : Pramod Chathampally, V. Mary Amala Bai |
10.14445/23488549/IJECE-V11I8P123
How to Cite?
Pramod Chathampally, V. Mary Amala Bai, "Beyond Pixels: Exploring Deep Learning Methods for Image Forgery Detection," SSRG International Journal of Electronics and Communication Engineering, vol. 11, no. 8, pp. 227-243, 2024. Crossref, https://doi.org/10.14445/23488549/IJECE-V11I8P123
Abstract:
Image forgery detection is a critical task in digital forensics, aiming to identify manipulated images to maintain trust and authenticity in digital content. Conventional techniques for identifying image forgeries often rely on handcrafted attributes and heuristics, which have limitations in detecting sophisticated forgeries. The capacity of deep learning algorithms to automatically extract pertinent features from data has made them a promising solution to this problem in recent years. The effectiveness of Convolutional Neural Networks (CNNs), a type of deep learning, in identifying image forgeries is investigated in this paper. The proposed research begins by collecting a dataset from the CASIA V2, comprising authentic and tampered images. Initially, a custom CNN model is constructed and trained on the dataset to establish a baseline performance. Subsequently, transfer learning using the MobileNet V2 architecture pretrained on the ImageNet dataset and is applied to leverage its feature extraction capabilities. However, the MobileNet V2 model demonstrates suboptimal accuracy before finetuning, prompting further enhancement. To improve the MobileNet V2 model’s efficiency, fine-tuning is employed at epoch 25, resulting in a notable accuracy increase to 94.14%. Compared to the baseline CNN model (93.98% accuracy) and the initial MobileNet V2 model (77.85% accuracy), fine-tuning significantly enhances the model’s efficiency in identifying image forgeries. The proposed methodology showcases the potential of deep learning in image forgery detection, offering improved accuracy and robustness in identifying manipulated digital content.
Keywords:
Image forgery, Authentic, Tampered, Transfer learning, CASIA V2, Compression error analysis.
References:
[1] Bin Xiao et al., “Image Splicing Forgery Detection Combining Coarse to Refined Convolutional Neural Network and Adaptive Clustering,” Information Sciences, vol. 511, pp. 172-191, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Yue Wu, Wael AbdAlmageed, and Premkumar Natarajan, “ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries with Anomalous Features,” 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, pp. 9535-9544, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Kalyani Dhananjay Kadam, Swati Ahirrao, and Ketan Kotecha, “Multiple Image Splicing Dataset (MISD): A Dataset for Multiple Splicing,” Data, vol. 6, no. 10, pp. 1-12, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Mohamed A. Elaskily et al., “Deep Learning Based Algorithm (ConvLSTM) for Copy Move Forgery Detection,” Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4385-4405, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Beste Ustubioglu et al., “A New Copy Move Forgery Detection Technique with Automatic Threshold Determination,” AEU - International Journal of Electronics and Communications, vol. 70, no. 8, pp. 1076-1087, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Babak Mahdian, and Stanislav Saic, “A Bibliography on Blind Methods for Identifying Image Forgery,” Signal Processing: Image Communication, vol. 25, no. 6, pp. 389-399, 2010.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Hany Farid, “Image Forgery Detection,” IEEE Signal Processing Magazine, vol. 26, no. 2, pp. 16-25, 2009.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Tran Van Lanh et al., “A Survey on Digital Camera Image Forensic Methods,” 2007 IEEE International Conference on Multimedia and Expo, Beijing, China, pp. 16-19, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Prajakta Kubal, Vanita Mane, and Namita Pulgam, “Image Manipulation Detection Using Error Level Analysis and Deep Learning,” International Journal of Intelligent Systems and Applications in Engineering, vol. 11, no. 4, pp. 91-99, 2023.
[Google Scholar] [Publisher Link]
[10] Shobhit Tyagi, and Divakar Yadav, “A Detailed Analysis of Image and Video Forgery Detection Techniques,” The Visual Computer, vol. 39, pp. 813-833, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Syed Sadaf Ali et al., “Image Forgery Detection Using Deep Learning by Recompressing Images,” Electronics, vol. 11, no. 3, pp. 1-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Emad Ul Haq Qazi, Tanveer Zia, and Abdulrazaq Almorjan, “Deep Learning-Based Digital Image Forgery Detection System,” Applied Sciences, vol. 12, no. 6, pp. 1-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Davide Alessandro Coccomini et al., “Cross-Forgery Analysis of Vision Transformers and CNNs for Deepfake Image Detection,” Proceedings of the 1st International Workshop on Multimedia AI against Disinformation, Newark NJ USA, pp. 52-58, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Haiwei Wu et al., “Robust Image Forgery Detection over Online Social Network Shared Images,” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA, pp. 13430-13439, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Amit Doegar et al., “Image Forgery Detection Based on Fusion of Lightweight Deep Learning Models,” Turkish Journal of Electrical Engineering and Computer Sciences, vol. 29, no. 4, pp. 1978-1993, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Sumaira Bibi et al., “Digital Image Forgery Detection Using Deep Autoencoder and CNN Features,” Human-Centric Computing and Information Sciences, vol. 11, no. 32, pp. 1-17, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Yohanna Rodriguez-Ortega, Dora M. Ballesteros, and Diego Renza, “Copy-Move Forgery Detection (CMFD) Using Deep Learning for Image and Video Forensics,” Journal of Imaging, vol. 7, no. 3, pp. 1-16, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Abhishek, and Neeru Jindal, “Copy Move and Splicing Forgery Detection Using Deep Convolution Neural Network, and Semantic Segmentation,” Multimedia Tools and Applications, vol. 80, pp. 3571-3599, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Shilpa Dua, Jyotsna Singh, and Harish Parthasarathy, “Image Forgery Detection Based on Statistical Features of Block DCT Coefficients,” Procedia Computer Science, vol. 171, pp. 369-378, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Boubacar Diallo et al., “Robust Forgery Detection for Compressed Images Using CNN Supervision,” Forensic Science International: Reports, vol. 2, pp. 1-11, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Lingzhi Li et al., “Face X-Ray for More General Face Forgery Detection,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, pp. 5000-5009, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Divyansh Goel, CASIA 2.0 Image Tampering Detection Dataset, Kaggle, [Online]. Available: https://www.kaggle.com/datasets/divg07/casia-20-image-tampering-detection-dataset
[23] Neal Krawetz, “A Picture’s Worth... Digital Image Analysis and Forensics Version 2,” Hacker Factor Solutions, pp. 1-43, 2007.
[Google Scholar] [Publisher Link]
[24] Rikiya Yamashita et al., “Convolutional Neural Networks: An Overview and Application in Radiology,” Insights into Imaging, vol. 9, pp. 611-629, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Ying Zhang et al., “Image Region Forgery Detection: A Deep Learning Approach,” Proceedings of the Singapore Cyber-Security Conference, Cryptology and Information Security Series, vol. 14, pp. 1-11, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Wina Permana Sari, and Hisyam Fahmi, “The Effect of Error Level Analysis on The Image Forgery Detection Using Deep Learning,” KINETIK: Game Technology, Information System, Computer Network, Computing, Electronics, and Control, vol. 6, no. 3, pp. 187-194, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Amit Doegar, Maitreyee Dutta, and Gaurav Kumar, “Image Forgery Detection Using Google Net and Random Forest Machine Learning Algorithm,” Journal of University of Shanghai for Science and Technology, vol. 22, no. 12, pp. 1271-1278, 2020.
[CrossRef] [Google Scholar] [Publisher Link]