Combining Deep Learning and Traditional Methods for Effective Copy-Move Forgery Detection in Digital Images

Copy-move forgery is a type of image tampering that involves copying a portion of an image and pasting it to another part of the same image with the intention of deceiving the viewer. In recent years, many approaches have been proposed to detect copy-move forgery, including those based on local binary patterns (LBP). In this paper, we perform a comprehensive evaluation of LBP-based methods for copy-move forgery detection using a dataset of 50 digital images. We compare the performance of four LBP-based methods, namely LBP, SIFT and SURF using metrics such as accuracy, precision, recall, and F1-score. Our results show that LBP outperforms the other methods in terms of accuracy and F1-score, while SIFT has the highest precision and recall. We also investigate the effect of various parameters, such as patch size and threshold values, on the performance of LBP. Our study provides valuable insights into the strengths and weaknesses of LBP-based methods for copy-move forgery detection, which can guide future research in this area. This study evaluates the performance of Local Binary Patterns (LBP) for detecting copy-move forgery in digital images. LBP is a widely used feature extraction technique in image processing and has been applied to various computer vision tasks, including forgery detection. The comparative study involves analyzing the accuracy, precision, recall, and F1-score of LBP and other popular forgery detection techniques, including SIFT and SURF, using a dataset of 50 digital images. The results show that LBP performs better than the other techniques, achieving an accuracy of 96.6%, precision of 94.0%, recall of 100%, and F1-score of 96.9%. This study provides useful insights for researchers and practitioners in the field of forgery detection, particularly for those interested in using LBP as a feature extraction technique.

Copy-move forgery is one of the most commonly used forgeries in digital images, in which some regions of a digital image is coped and pasted to another part of the image with the aim of concealing certain features or objects in the original images. In this paper, a novel copy-move forgery detection scheme in digital images is proposed. It treats the digital images as high dimensional data and uses the intrinsic dimension estimation method first to segment the image and then identify the copy-move forgeries in the image regions with the same texture. Experimental results show that the proposed method is efficient and robust to retouching and many other operations, such as lossy compression, blurring, filtering, etc.

The increasing accessibility of digital imaging technology has led to a rise in image forgery, raising concerns about digital content authenticity in forensic and security domains. Copy-move forgery is one the most prevalent and challenging forgery techniques due to its seamless manipulation. We propose a novel passive CMFD (CMFD) approach that leverages a modified Kirsch (mKirsch) edge detector and a modified SIFT-based descriptor (DivSIFT) to accurately identify and localize copy-move forgery (CMF). The mKirsch edge detector enhances edge detection by selectively deleting specific masks, improving keypoint extraction and feature matching. We used MICC-F220, CoMoFoD, and MICC-F8Multi datasets to measure the performance of the new method, under challenging conditions such as rotation, scaling, JPEG compression, and multiple forgeries. The results show that mKirsch-enhanced detection outperforms compared to conventional Kirsch-based methods. Notably, methods with deleted masks (WW_NW and NE_SE) achieved a True Positive Rate (TPR) of 90.91%, precision 100%, and an F-measure of 95.24%. Robust against rotation and scaling attacks, achieving a TPR of up to 96.97% with zero false positives. Additionally, the method is computationally efficient, with an execution time of 2.74 seconds, making it suitable for real-world applications. These findings establish the mKirsch-based CMFD as a highly accurate and efficient solution for image forgery detection in digital images.

In this paper, an adaptive method for copy-move forgery detection and localization in digital images is proposed. The method employs wavelet transform with non constant Q factor and characterizes image pixels through the multiscale behavior of corresponding wavelet coefficients. The detection of forged regions is then performed by considering similar those pixels having the same multiscale behavior. The method is pointwise and the length of pixel features vector is image dependent, allowing for a more precise and fast detection of forged regions. The qualitative and quantitative evaluation of the experimental results reveals that the proposed method outperforms some existing transform-based methods in terms of performance and execution time.

In recent years, digital image forgery detection has become an active research area due to the advancement of photo editing software. This paper focuses on passive forgery detection on images tampered using copy move technique, better known as Copy Move Forgery Detection (CMFD). A CMFD technique consisting of oriented Features from Accelerated Segment Test and rotated Binary Robust Independent Elementary Features (Oriented FAST and rotated BRIEF) as the feature extraction method and 2 Nearest Neighbour (2NN) with Hierarchical Agglomerative Clustering (HAC) as the feature matching method is proposed. Evaluation of the proposed CMFD technique was performed on images that underwent various geometrical attacks. With the proposed technique, an overall accuracy rate of 84.33% and 82.79% are obtained for evaluation carried out with images from the MICC-F600 and MICC-F2000 databases. Forgery detection achieved True Positive Rate of more than 91% for tampered images with object translation, different degree of rotation and enlargement.

With the rapid advancement of image processing techniques, digital image forgery detection has emerged as a critical research area in information forensics. This paper proposes a novel deep learning model based on Multi-view Multi-dimensional Forgery Detection Networks (MMFD-Net), designed to simultaneously determine whether an image has been tampered with and precisely localize the forged regions. By integrating a Multi-stream Edge Feature Learning module with a Multi-dimensional Information Fusion module, MMFD-Net employs joint supervised learning to extract semantics-agnostic forgery features, thereby enhancing both detection performance and model generalization. Extensive experiments demonstrate that MMFD-Net achieves state-of-the-art results on multiple public datasets, excelling in both pixel-level localization and image-level classification tasks, while maintaining robust performance in complex scenarios.

Digital images are widely used nowadays and there are powerful and convenient software tools for various types of image processing and enhancement. Unfortunately, these tools also facilitate for easy digital image forgery. In this paper we present an algorithm for digital image forgery detection that deals with the situation when some object, together with its shadow, is copied and pasted to some other location in the same or different image. Algorithm is based on the property that shadows do not change the texture of the underlaying surface. Our algorithms uses local binary patterns from shadow and adjacent non-shadow regions and features extracted from their histograms where energy and entropy proved to be the most discriminative. The proposed method was tested on some benchmark forged images and compared with other approaches from literature where it proved to be successful in detection of this type of forgery.

A passive blind digital image forgery detection method was proposed in this paper. Basic defocus model shows that image patches with similar distances to the lens have similar blur kernel sizes. This consistency is broken in image forgery as the result of possible blurring and different imaging conditions. Our forgery detection technique uses local blur estimation at each edge pixels to exposes the defocus blur inconsistency. Experiment results of tampered images from real law cases show the effectiveness of our technique.

Statistical feature-based pattern recognition approach has been proved to be useful in digital image forgery detection. In this paper, we first present this approach adopted in Phase 1 of the first IEEE IFS-TC Image Forensics Challenge, in which the task is to classify the tampered images from the original ones, together with the experimental results. Several different kinds of statistical features and their combinations have been tested. Finally, we have chosen to use co-occurrence matrices calculated from the rich model of noise residual images. Furthermore we have selected a subset of the rich model to further enhance the testing accuracy by about 2 % so as to reach a high detection accuracy of 93.7 %. In Phase 2 of the competition, the task is to localize the tampered regions by identifying the tampered pixels. For this purpose, we have introduced the Hamming distance of Local Binary Patterns as similarity measure to tackle the tampering without post-processing on copy-moved regions. The PatchMatch algorithm has been adopted as an efficient search algorithm for block-matching. We have also applied a simple usage of the scale-invariant feature transform (SIFT) when other kinds of processing such as rotation and scaling have been performed on the copy-moved region. The achieved f-score in identifying tampered pixels is 0.267. In summary, some success has been achieved apparently. However, much more efforts are called for to move image tampering detection ahead.

The current deep learning methods for copy-move forgery detection (CMFD) are mostly based on deep convolutional neural networks, which frequently discard a large amount of detail information throughout convolutional feature extraction and have poor long-range information extraction capabilities. The Transformer structure is adept at modeling global context information, but the patch-wise self-attention calculation still neglects the extraction of details in local regions that have been tampered with. A local-information-refined dual-branch network, LBRT (Local Branch Refinement Transformer), is designed in this study. It performs Transformer encoding on the global patches segmented from the image and local patches re-segmented from the global patches using a global modeling branch and a local refinement branch, respectively. The self-attention features from both branches are precisely fused, and the fused feature map is then up-sampled and decoded. Therefore, LBRT considers both global semantic information modeling and local detail information refinement. The experimental results show that LBRT outperforms several state-of-the-art CMFD methods on the USCISI dataset, CASIA CMFD dataset, and DEFACTO CMFD dataset.

ELA-Conv: Forgery Detection in Digital Images Based on ELA and CNN

The progression and diversification of methodologies in digital image forensics and manipulation detection are evident in existing work, presenting innovative techniques tailored to specific aspects of forgery detection. These methodologies encompass various challenges such as copy-move forgery, manipulation detection, demosaicing artifacts, image splicing, edge detection, and compression analysis. Together, these studies exemplify ongoing efforts to enhance the accuracy, efficiency, and reliability of forgery detection methods within the constantly evolving landscape of digital image manipulation. The proposed image forgery detection system represents a pioneering advancement, surpassing the limitations of conventional methodologies. It integrates a comprehensive array of detection techniques, including copy-move analysis, color filter array scrutiny, noise variance inconsistency detection, and double JPEG artifact identification. These techniques are bolstered by Convolutional Neural Networks (CNNs). This innovative fusion not only addresses the limitations inherent in existing systems but also signifies a significant leap forward in functionality and efficacy. By synergistically leveraging the strengths of each detection method within a CNN framework, our approach promises heightened accuracy, robustness, and adaptability in discerning even the most sophisticated forms of image tampering. Keywords— Image Tampering, Double JPEG Compression, Demosaicing Artifacts, Recursive Edge Detection, Copy-Move Forgery Detection.

ObjectivesThis study aimed to investigate the accuracy of deep learning algorithms to diagnose tooth caries and classify the extension and location of dental caries in cone beam computed tomography (CBCT) images. To the best of our knowledge, this is the first study to evaluate the application of deep learning for dental caries in CBCT images. MethodsThe CBCT image dataset comprised 382 molar teeth with caries and 403 noncarious molar cases. The dataset was divided into a development set for training and validation and test set. Three images were obtained for each case, including axial, sagittal, and coronal. The test dataset was provided to a multiple-input convolutional neural network (CNN). The network made predictions regarding the presence or absence of dental decay and classified the lesions according to their depths and types for the provided samples. Accuracy, sensitivity, specificity, and F1 score values were measured for dental caries detection and classification. ResultsThe diagnostic accuracy, sensitivity, specificity, and F1 score for caries detection in carious molar teeth were 95.3%, 92.1%, 96.3%, and 93.2%, respectively, and for noncarious molar teeth were 94.8%, 94.3%, 95.8%, and 94.6%. The CNN network showed high sensitivity, specificity, and accuracy in classifying caries extensions and locations. ConclusionsThis research demonstrates that deep learning models can accurately identify dental caries and classify their depths and types with high accuracy, sensitivity, and specificity. The successful application of deep learning in this field will undoubtedly assist dental practitioners and patients in improving diagnostic and treatment planning in dentistry. Clinical significance: This study showed that deep learning can accurately detect and classify dental caries. Deep learning can provide dental caries detection accurately. Considering the shortage of dentists in certain areas, using CNNs can lead to broader geographic coverage in detecting dental caries.

Capturing images has been increasingly popular in recent years, owing to the widespread availability of cameras. Images are essential in our daily lives because they contain a wealth of information, and it is often required to enhance images to obtain additional information. A variety of tools are available to improve image quality; nevertheless, they are also frequently used to falsify images, resulting in the spread of misinformation. This increases the severity and frequency of image forgeries, which is now a major source of concern. Numerous traditional techniques have been developed over time to detect image forgeries. In recent years, convolutional neural networks (CNNs) have received much attention, and CNN has also influenced the field of image forgery detection. However, most image forgery techniques based on CNN that exist in the literature are limited to detecting a specific type of forgery (either image splicing or copy-move). As a result, a technique capable of efficiently and accurately detecting the presence of unseen forgeries in an image is required. In this paper, we introduce a robust deep learning based system for identifying image forgeries in the context of double image compression. The difference between an image’s original and recompressed versions is used to train our model. The proposed model is lightweight, and its performance demonstrates that it is faster than state-of-the-art approaches. The experiment results are encouraging, with an overall validation accuracy of 92.23%.

Deep convolutional neural networks (DCNNs) for diagnosis of disease on chest radiographs (CXR) have been shown to be biased against males or females if the datasets used to train them have unbalanced sex representation. Prior work has suggested that DCNNs can predict sex on CXR, which could aid forensic evaluations, but also be a source of bias. To (1) evaluate the performance of DCNNs for predicting sex across different datasets and architectures and (2) evaluate visual biomarkers used by DCNNs to predict sex on CXRs. Chest radiographs were obtained from the Stanford CheXPert and NIH Chest XRay14 datasets which comprised of 224,316 and 112,120 CXRs, respectively. To control for dataset size and class imbalance, random undersampling was used to reduce each dataset to 97,560 images that were balanced for sex. Each dataset was randomly split into training (70%), validation (10%), and test (20%) sets. Four DCNN architectures pre-trained on ImageNet were used for transfer learning. DCNNs were externally validated using a test set from the opposing dataset. Performance was evaluated using area under the receiver operating characteristic curve (AUC). Class activation mapping (CAM) was used to generate heatmaps visualizing the regions contributing to the DCNN's prediction. On the internal test set, DCNNs achieved AUROCs ranging from 0.98 to 0.99. On external validation, the models reached peak cross-dataset performance of 0.94 for the VGG19-Stanford model and 0.95 for the InceptionV3-NIH model. Heatmaps highlighted similar regions of attention between model architectures and datasets, localizing to the mediastinal and upper rib regions, as well as to the lower chest/diaphragmatic regions. DCNNs trained on two large CXR datasets accurately predicted sex on internal and external test data with similar heatmap localizations across DCNN architectures and datasets. These findings support the notion that DCNNs can leverage imaging biomarkers to predict sex and potentially confound the accurate prediction of disease on CXRs and contribute to biased models. On the other hand, these DCNNs can be beneficial to emergency radiologists for forensic evaluations and identifying patient sex for patients whose identities are unknown, such as in acute trauma.