Image Forgery Detection Using Tamper-Guided Dual Self-Attention Network with Multiresolution Hybrid Feature

Li, Fengyong; Pei, Zhenjia; Wei, Weimin; Li, Jing; Qin, Chuan

doi:10.1155/2022/1090307

Cited by 6 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, researchers have tried to identify copy-move and splicing frauds concurrently. A new approach uses a fully convolutional network with multi-resolution hybrid features [8]. Tamper-guided dual self-attention module in this network distinguishes tampered regions from unaffected ones.…”

Section: Related Workmentioning

confidence: 99%

Detecting Digital Image Forgeries with Copy-Move and Splicing Image Analysis using Deep Learning Techniques

Timothy,

Santra

2024

ijacsa

View full text Add to dashboard Cite

The proliferation of digitally altered images across social media platforms has escalated the urgency for robust image forgery detection systems. Traditional detection methodologies, while varied, often fall short in addressing the multifaceted nature of image forgeries in the digital landscape. Recognizing the need for advanced solutions, this paper introduces a novel deep-learning approach that leverages the architectural strengths of GNNs, CNNs, VGG16, MobileNet, and ResNet50. Our method uniquely integrates these architectures to effectively detect and analyze multiple types of image forgeries, including image splicing and copy-move forgeries. This approach is groundbreaking as it adapts these networks to focus on identifying discrepancies in the compression quality between forged and original image regions. By examining the differences between the original and compressed image versions, our model constructs a feature-rich representation, which is then analyzed by a tailored deep-learning network. This network has been enhanced by removing its original classifier and implementing a new one specifically designed for binary forgery classification. Very few researchers have explored the application of deep learning techniques in copy-move and splice image analysis for detecting digital image forgeries, making our work particularly significant. A comprehensive comparative analysis with pre-trained models underscores the superiority of our method, with the GNN model achieving an impressive accuracy of 98.54 percent on the CASIA V1 dataset. This not only sets a new benchmark in the field but also highlights the efficiency of our model, which benefits from reduced training parameters and accelerated training times.

show abstract

Section: Related Workmentioning

confidence: 99%

Detecting Digital Image Forgeries with Copy-Move and Splicing Image Analysis using Deep Learning Techniques

Timothy,

Santra

2024

ijacsa

View full text Add to dashboard Cite

show abstract

“…They introduce a novel approach incorporating color illumination, DL techniques, and semantic segmentation to achieve high accuracy in forgery detection and localization. They use various feature extraction techniques, including multi-radius PCET, autoencoder, KAZE interest point detector, HoG, HOGG, and CNN, to extract features relevant to forgery detection, and they used SVM, KNN, Li et al[28] combine visual and noise inconsistency artifacts to identify tampered images efficiently. Their proposed technique uses an endto-end fully CNN with multiresolution hybrid features from RGB and noise streams and a tamper-guided dual self-attention module for accurate tampered area focus.…”

mentioning

confidence: 99%

A Holistic Approach to Image Forensics: Integrating Image Metadata Analysis and ELA with CNN and MLP for Image Forgery.

NFOR,

Armand,

Kim

2024

Preprint

View full text Add to dashboard Cite

In the face of the escalating use of image editing tools and the rapid progress of AI technology, the demand for robust image forgery detection techniques has become more pressing. This study introduces a comprehensive approach to image forgery detection, which integrates image metadata analysis and Error Level Analysis (ELA) with deep learning, specifically Convolutional Neural Networks (CNNs) and Multilayer Perceptrons (MLPs). Using transfer learning, we evaluated seven pre-trained models alongside a custom innovative CNN model using the CASIAV2 dataset for binary classification of authentic and forged images. Metadata was extracted from all the images, and unique feature images were generated and used as inputs for all models by exploiting the differences between compressed and original images through ELA. Among the pre-trained models, VGG16, Xception, ResNet101, and MobileNetV2 demonstrated superior detection accuracy compared to the state-of-the-art models. However, our custom model surpassed six others, achieving an accuracy score of 99.08% with efficient training parameters. This study's results indicate significant advancements in image forgery detection, which has practical applications in forensic tools, automated detection systems, and educational purposes. An enhanced accuracy can benefit industries reliant on image authenticity, improve digital security protocols, protect intellectual property, and influence legal standards. Additionally, the methodology could extend to other domains like document and video authentication.

show abstract

Enhancing Digital Image Forgery Detection Using Transfer Learning

Khalil,

Ghalwash,

Elsayed

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Nowadays, digital images are a main source of shared information in social media. Meanwhile, malicious software can forge such images for fake information. So, it's crucial to identify these forgeries. This problem was tackled in the literature by various digital image forgery detection techniques. But most of these techniques are tied to detecting only one type of forgery, such as image splicing or copy-move that is not applied in real life. This paper proposes an approach, to enhance digital image forgery detection using deep learning techniques via transfer learning to uncover two types of image forgery at the same time, The proposed technique relies on discovering the compressed quality of the forged area, which normally differs from the compressed quality of the rest of the image. A deep learning-based model is proposed to detect forgery in digital images, by calculating the difference between the original image and its compressed version, to produce a featured image as an input to the pre-trained model to train the model after removing its classifier and adding a new fine-tuned classifier. A comparison between eight different pre-trained models adapted for binary classification is done. The experimental results show that applying the technique using the adapted eight different pre-trained models outperforms the state-of-the-art methods after comparing it with the resulting evaluation metrics, charts, and graphs. Moreover, the results show that using the technique with the pre-trained model MobileNetV2 has the highest detection accuracy rate (around 95%) with fewer training parameters, leading to faster training time.

show abstract

Image Forgery Detection Using Tamper-Guided Dual Self-Attention Network with Multiresolution Hybrid Feature

Cited by 6 publications

References 37 publications

Detecting Digital Image Forgeries with Copy-Move and Splicing Image Analysis using Deep Learning Techniques

Detecting Digital Image Forgeries with Copy-Move and Splicing Image Analysis using Deep Learning Techniques

A Holistic Approach to Image Forensics: Integrating Image Metadata Analysis and ELA with CNN and MLP for Image Forgery.

Enhancing Digital Image Forgery Detection Using Transfer Learning

Contact Info

Product

Resources

About