Digital Manipulation Research Articles

Forgered Image Perception System using CNN Algorithms

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.

Fishboneviz: Enhancing the availability of zooarchaeological fish reference collections through an open access 3D database

Fish remains are a common component of coastal and inland archaeological assemblages from Australia and the Pacific Islands. Physical reference collections are the primary tool that researchers use to taxonomically identify fish bones but given the high biodiversity of fishes in this region, collections are frequently not complete at the genus- and species-level. Adjunct resources, such as online photograph repositories of fish bone reference collections and illustrated technical guides, provide useful supplementary aids. However, such 2D photographs and illustrations offer fixed perspectives of the bone and do not allow for ready manipulation and detailed examination of the specimen. Here, we introduce Fishboneviz, the first open access 3D fish bone reference collection, which was developed to reduce inequitable access to physical reference collections in line with the FAIR principles of data management (findable, accessible, interoperable, reusable). A ‘best-practice’ methodology was established to facilitate fast and effective segmentation (i.e. isolate regions of interest such as elements) of fish Computed Tomography (CT) scans. This way, fish bone elements of interest were segmented to allow digital manipulation and viewing of the complete element. To examine the effectiveness of the approach, image segmentation procedures were applied to a representative sample of 10 bone elements per fish: dentary, premaxilla, maxilla, articular, quadrate, hyomandibular, opercle, preopercle, last precaudal vertebra, and first caudal vertebra. For species within the family Labridae, the three pharyngeal grinding plates were also segmented. These elements were selected as they represent different regions of the skeleton, variable morphologies, and commonly recovered elements from archaeological sites. To date, the collection contains the skeletal elements of 26 fish species (18 families). In the future, it is hoped that the collection will be further expanded by a broader network of interested collaborators to ensure it grows according to the changing needs of research and teaching communities.

Deep Neural Networks for Multimedia Forensics: Detection of Deepfake Content

Deep Neural Networks for Multimedia Forensics: Detection of Deepfake Content

Unveiling Digital Document Manipulation: A Case Study in Forensic Examination

The ease of creating digital documents due to today’s technological advancements has led to a surge in white- collar crimes involving forgery and manipulation. Detecting these digital document alterations presents a unique set of challenges to forensic investigators. The present study aims to uncover a forgery in a digitally manipulated document and demonstrate the successful detection of digital document alterations. This study examines a real- life case to explore the methods utilized by forensic experts in detecting and evaluating document forgery. It highlights the crucial role of digital forensic analysis in contemporary investigative procedures. This research emphasises the significance of understanding document forgery and the need for robust forensic techniques in detecting digital manipulation. Ensuring the integrity of digital documents is crucial in maintaining trust and reliability in today’s information-driven era.

DEEP FACE LIVE

In Deep Face Live is a cutting-edge application designed to revolutionize the field of digital media manipulation by enabling seamless face replacement in videos. Leveraging state-of-the-art deep learning models and algorithms, Deep Face Live empowers users to create compelling and realistic content with unprecedented ease and efficiency. This research paper presents a comprehensive overview of Deep Face Live, including its architecture, functionalities, and applications. We discuss the methodology employed in the development of Deep Face Live, highlighting key design decisions, algorithmic approaches, and technical considerations. Furthermore, we evaluate the performance of Deep Face Live through qualitative and quantitative analysis, demonstrating its effectiveness in generating high-quality deep fake content. Additionally, we address ethical considerations surrounding the use of Deep Face Live and propose strategies for responsible deployment and usage. Overall, this research paper aims to contribute to the advancement of digital media technology and foster informed discourse on the ethical implications of Deep Face Live. Index Terms : Deep learning, Deep Fake, review, Deep Fake generation, Deep Fake creation

Image Copyright Protection Based on Blockchain Technology Review

On a daily basis, a significant number of individuals distribute several photos and videos that have been marginally modified from the original material produced by copyright owners, such as photographers, graphic designers, and video producers. Individuals that infringe upon the rights of others, lacking the legal authority to access multimedia content, employ various digital image and picture manipulation techniques, it involves converting to gray scale, trimming, rotating, contracting the frame, and adjusting the background speed, to modify said content. Blockchain technology obviates the necessity of an intermediary, hence circumventing the possibility of a singular point of failure. Infractions to copyright poses a significant barrier to protecting commercial image and video information. The IPFS blockchain technology offers on-chain preservation for copyright information and off-chain storing for distinct multimedia files. The enhanced perceptual hashing algorithm significantly enhances the precision of identifying connections to identify digital image piracy. The photographers and designers that submit their photographs on websites are experiencing significant dissatisfaction due to a prevalent practice in which others attempt to claim credit and profit from the initial creator's effort.

Efficient Deepfake Audio Detection Using Spectro-Temporal Analysis and Deep Learning

With the advancement of deepfake technology, particularly in the audio domain, there is an imperative need for robust detection mechanisms to maintain digital security and integrity. Our research presents an "Enhanced Deepfake Audio Detection with Spectro-Temporal Deep Learning Approach," aimed at addressing the escalating challenge of distinguishing genuine audio from sophisticated deepfake manipulations. Our methodology leverages the ADD2022 dataset, which encompasses a wide range of audio clips, to train and evaluate a novel deep learning model. The core of our approach consists of a meticulous data preprocessing phase, where audio samples are resampled, normalized, and subjected to silence removal to ensure uniformity and enhance model input quality. Our deep learning model architecture innovatively combines Convolutional Neural Networks (CNNs) for extracting spectral features and Recurrent Neural Networks (RNNs) or Long Short-Term Memory (LSTM) networks for analyzing temporal dynamics. This hybrid model is adept at identifying the nuanced anomalies characteristic of deepfake audios. The model undergoes rigorous training and optimization, with performance evaluated against key metrics such as accuracy, precision, F1, recall score, and the Equal Error Rate (EER). Our findings exhibit the Enhanced Deepfake Audio Detection model's superior capability to discern deepfake audio, highlighting its potential as a pivotal tool in the fight against digital audio manipulation. This research contributes significantly to the field of digital forensics, offering a scalable and effective solution for ensuring the authenticity of audio content in an era dominated by deepfake technology.

Digital holographic content manipulation for wide-angle holographic near-eye displays

In recent years, the development of holographic near-eye displays (HNED) has surpassed the progress of digital hologram recording systems, especially in terms of wide-angle viewing capabilities. Thus, there is capture-display parameters incompatibility, which makes it impossible to reconstruct recorded objects in wide-angle display. This paper presents a complete imaging chain extending the available content for wide-angle HNED of pupil and non-pupil configuration with narrow-angle digital holograms of real objects. To this end, a new framework based on the phase-space approach is proposed that includes a set of affine transformations required to account for all differences in capture-display cases. The developed method allows free manipulation of the geometry of reconstructed objects, including axial and lateral positioning and size scaling. At the same time, it has a low computational effort. The presented work is supported with non-paraxial formulas developed using the phase-space approach, enabling accurate tracing of the holographic signal, its reconstruction, and measuring appearing deformations. The applicability of the proposed hologram manipulation method is proven with experimental results of digital hologram reconstruction in wide-angle HNED.

Survey on Deep Fake Detection using Deep Learning

The rise of deep fake technology poses significant challenges to the authenticity and integrity of visual content on digital platforms. This paper presents the development of a web application aimed at detecting deep fake images and videos through the implementation of advanced deep learning models. Leveraging research in the field of deep fake detection, this application integrates state-of-the-art algorithm like CNNs for accurate classification tasks. Key features of the application include a user-friendly interface allowing for the upload and analysis of images and videos, leveraging the trained models to detect potential deep fakes. Additionally, the web application incorporates a sophisticated detection methods to enhance accuracy and reliability. The deployment of the web application on popular platforms aims to provide users with a tool to verify the authenticity of visual content, thereby mitigating the potential negative consequences of deep fake manipulation. Continuous monitoring, updates, and adaptation to emerging deep fake techniques are prioritized to ensure the web applications effectiveness and relevance in an ever-evolving landscape of digital content manipulation

Influence of education level and gender of dental students on perception of dental aesthetics

BackgroundFulfilling aesthetic norms is an imperative of modern society. Accordingly, aesthetics has found its important role in dental medicine. The aim was to study whether there is a difference in the perception of tooth shade change and anatomical variations of anterior maxillary teeth among dental students depending on the level of education, gender, and experience in assisting in a dental office.MethodsThe research was conducted among dental students via the Google Forms platform. Students evaluated the aesthetic acceptability of a smile on the unaltered, reference photograph (RP) and on the other 28 photographs in which the tooth shade (lighter and darker teeth), length (shortened central incisors, elongated, and beveled lateral incisors, elongated canines) and position (atypical and typical rotations and diastemas) of the upper maxillary teeth were altered by digital manipulation. The 1–10 assessment scale was used. Statistical analysis comprised one-way Kolmogorov-Smirnov test, X2 tests, t-test, and 3-way MANOVA.ResultsIn the research 208 students participated, 113 were preclinical students, 104 assisted in dental office and 175 were females. There were no significant effects of gender regarding length and position modifications (p > .05), while women were stricter in evaluation of lighter shade (p < .05). Clinical students were stricter in evaluating all manipulated photographs (MP) except those towards darker shade modifications, while assisting in a dental office or having a dentist in close family showed no significant effects (p > .05).ConclusionsWith a higher level of targeted academic education, dental students sharpen their ability to notice deviations from the aesthetic norms of a smile.

Unveiling Digital Manipulation and Persuasion in e-Commerce: A Systematic Literature Review of Dark Patterns and Digital Nudging

Dark patterns and digital nudging are persuasive techniques that influence customer decision-making in the digital world. This study analyzed 80 articles published in reputed peer-reviewed journals and international conferences using the Theory-Context-Characteristics-Methods (TCCM) framework. The paper offers five significant contributions. First, it adds to the theoretical understanding of the area. Second, the study identifies various contexts in which dark patterns and digital nudging are employed. Third, it elucidates the characteristics (key variables) of dark patterns and digital nudging. Fourth, the study determines the methodologies used in prior studies in the field. Lastly, the paper proposes future research areas in the field. By presenting a comprehensive overview of the research in the field, the study could stimulate scholarly research and serve as a valuable base for future researchers. Furthermore, insights into dark patterns and digital nudging will educate online customers and allow managers to devise strategies to mitigate customer pain points.

Os documentos do palimpsesto de Santa Sofia di Benevento (século XI) estudados através da técnica de “imagem multiespectral”

Although the palimpsests codes are not longer a rarity, the rescripti documents are still very rare. It was the practice of reusing parchments already written, then erased and subjected to a particular treatment, for the purpose of recycling for the preparation of a new document. The study in question examines the parchments of Santa Sofia di Benevento and the conspicuous dossier consisting of documents concerning the monastery of Sant’Angelo di Ariano. The membranes examined for this study are twenty-two (twenty-one have double palimpsests, while one is triple), for a total of forty-five documents, all relating to the eleventh century. The new multispectral imaging technique has allowed the identification of new occurrences relating to the sofiano parchment background, through the experimentation of multispectral digital photography and the digital manipulation of dissected images, able to bring out, in their clarity, the traces of the writings no longer visible either to the naked eye, or through the classic use of the Wood Lamp; this new photographic technique allows the isolation and separation of one or more levels of writing.

Digital manipulation and mass mobilization over the long run: evidence from Latin America

False information flows through social media are a worldwide phenomenon, affecting Latin America in particular, where more than 70% of internet users report growing concern regarding the circulation of fake news, especially through social media platforms. Although the academic literature has studied how governments may be incentivized to disseminate false information, it has concentrated in the electoral context mostly. In this sense, the impact of disinformation on mass mobilization patterns has been less explored, although it is relevant: manipulation through social media can reduce citizens' inclination to protest by directly influencing their preferences or beliefs about the government, acting as a preventive digital repression tool. In fact, governments have efficiently used it in the short run; however, it has not been established if this strategy could work in the long run. Hence, this article explores the long-term relationship between manipulation by the government through social media and the occurrence of mass mobilization to determine whether they are cointegrated. Drawing on the most recent literature on digital repression, manipulation, and mass mobilization and by applying a panel cointegration methodology in nine countries in Latin America (2000–2021), we find that the variables are cointegrated. In addition, we explore the causal (Granger) effect and find a unidirectional link that goes from social media manipulation to mass mobilization.

A EFFCIET DEEP FAKE FACE DETECTION USING DEEP INCEPTION NET LEARNING ALGORITHM

A Deep Fake Is Digital Manipulation Techniques That Use Deep Learning to Produce Deep Fake (Misleading) Images and Videos. Identifying Deep Fake Images Is the Most Difficult Part of Finding the Original. Due To the Increasing Reputation of Deep Fakes, Identifying Original Images and Videos Is More Crucial to Detect Manipulated Videos. This Paper Studies and Experiments with Different Methods That Can Be Used to Detect Fake and Real Images and Videos. The Convolutional Neural Network (Cnn) Algorithm Named Inception Net Has Been Used to Identify Deep Fakes. A Comparative Analysis Was Performed in This Work Based on Various Convolutional Networks. This Work Uses the Dataset from Kaggle With 401 Videos of Train Sample And 3745 Images Were Generated by Augmentation Process. The Results Were Evaluated with The Metrics Like Accuracy and Confusion Matrix. The Results of The Proposed Model Produces Better Results in Terms of Accuracy With 93% On Identifying Deep Fake Images and Videos.

Assessment framework for deepfake detection in real-world situations

Detecting digital face manipulation in images and video has attracted extensive attention due to the potential risk to public trust. To counteract the malicious usage of such techniques, deep learning-based deepfake detection methods have been employed and have exhibited remarkable performance. However, the performance of such detectors is often assessed on related benchmarks that hardly reflect real-world situations. For example, the impact of various image and video processing operations and typical workflow distortions on detection accuracy has not been systematically measured. In this paper, a more reliable assessment framework is proposed to evaluate the performance of learning-based deepfake detectors in more realistic settings. To the best of our acknowledgment, it is the first systematic assessment approach for deepfake detectors that not only reports the general performance under real-world conditions but also quantitatively measures their robustness toward different processing operations. To demonstrate the effectiveness and usage of the framework, extensive experiments and detailed analysis of four popular deepfake detection methods are further presented in this paper. In addition, a stochastic degradation-based data augmentation method driven by realistic processing operations is designed, which significantly improves the robustness of deepfake detectors.

Exploring the Effectiveness of Machine Learning Algorithms in Image Forgery Detection

This study investigates the efficacy of various machine learning algorithms for detecting image forgery, a prevalent issue in the realm of digital media manipulation. The research focuses on assessing the performance of these algorithms in accurately identifying instances of image tampering, aiming to contribute valuable insights to the field of digital forensics. The evaluation encompasses a diverse set of machine learning techniques, including but not limited to convolutional neural networks (CNNs), support vector machines (SVMs), and decision trees. Through rigorous experimentation and comparative analysis, the research aims to discern the strengths and limitations of each algorithm in the context of image forgery detection. The findings of this study hold significance for enhancing the capabilities of digital forensics tools, thereby aiding in the mitigation of fraudulent activities, and ensuring the integrity of visual content in the digital' domain.

Deepfake image detection and classification model using Bayesian deep learning with coronavirus herd immunity optimizer

&lt;p&gt;Deepfake images are combined media constructed from deep learning (DL) methods, usually Generative Adversarial Networks (GANs), to manipulate visual content, often giving rise to convincing and fabricating descriptions of scenes or people. The Bayesian machine learning (ML) model has made crucial strides over the past two decades, illustrating promise in diverse applications. In deepfake images, detection utilizes computer vision (CV) and ML to spot manipulated content by analyzing unique artefacts and patterns. Recent techniques utilize DL to train neural networks to discriminate between real and fake images, improving the fight against digital manipulation and preserving media integrity. These systems can efficiently detect subtle inconsistencies or anomalies specific to deepfake creations by learning from large datasets of both real and deepfake images. This enables the mitigation of fraudulent content and reliable detection in digital media. We introduce a new Coronavirus Herd Immunity Optimizer with a Deep Learning-based Deepfake Image Detection and Classification (CHIODL-DIDC) technique. The CHIODL-DIDC technique aimed to detect and classify the existence of fake images. To accomplish this, the CHIODL-DIDC technique initially used a median filtering (MF) based image filtering approach. Besides, the CHIODL-DIDC technique utilized the MobileNetv2 model for extracting feature vectors. Moreover, the hyperparameter tuning of the MobileNetv2 model was accomplished using the CHIO method. For deepfake image detection, the CHIODL-DIDC technique implements the deep belief network (DBN) model. Finally, the Bayesian optimization algorithm (BOA) was utilized to select the effectual hyperparameter of the DBN model. The CHIODL-DIDC method's empirical analysis was examined using a benchmark fake image dataset. The performance validation of the CHIODL-DIDC technique illustrated a superior accuracy value of 98.16% over other models under $ Acc{u}_{y} $ , $ Pre{c}_{n} $ , $ Rec{a}_{l} $ , $ {F}_{Score} $ , and MCC metrics.&lt;/p&gt;

Cyber vaccinator for image tamper resilient and recovery using invertible neural network

People frequently interact with their families, friends, and colleagues through Online Social Networks (OSNs). People post and share their photos in online communities and content-sharing sites. The problem addressed in this paper is the susceptibility of digital images to tampering, which compromises security and privacy. Traditional image forgery detection methods face challenges in reproducing original content after manipulation. This paper introduces an advanced Image Immunization System leveraging Invertible Neural Networks. The system, which comprises the cyber vaccinator, vaccine validator, forward pass for tamper detection, and backward pass for image self-recovery, aims to proactively immunize images against various attacks. The run-length encoding in the backward pass transforms hidden perturbations into information, facilitating the recovery of the authentic image. The middleware's expansion to multimodal content analysis, including videos and audio, provides a more comprehensive defense against digital manipulation within OSNs. These advancements reflect a commitment to robust security and holistic content integrity. The Cyber Vaccinator, using Invertible Neural Networks (INNs) for image tamper resilience and recovery, demonstrates significant effectiveness in detecting tampering and restoring images, providing a robust solution for maintaining image integrity. The Cyber Vaccinator uses an Invertible Neural Network (INN) to safeguard image integrity. It detects tampering by analyzing invariant features and responds with precise recovery methods. By continuously monitoring images, it ensures real-time tamper detection and efficient restoration, maintaining image authenticity through advanced neural network resilience and recovery techniques.

Philosophy Through Computation

We explore the possibility of teaching philosophy through the teaching of computer programming. It is pedagogically useful to use programming because it is extremely popular (especially due to the recent breakthroughs in machine learning), and it can provide a novel, interesting, and clear introduction to a variety of classic philosophical issues. We discuss two examples. The first is using programming to solve digital image manipulation tasks as a way of introducing and clarifying debates over external world skepticism. The second is programming a simulation of a simple 2-D cellular automaton as a way of introducing and clarifying debates over determinism and free will.

Surface acoustic wave digital microfluidics with surface wettability gradient.

This paper reports a digital microfluidic technology that combines surface wettability gradient and surface acoustic waves. The technology enables selection of the driven object, facilitating reactions among multiple droplets and improving the precision of droplet control. Octagonal patterns with a wetting gradient and orthogonally distributed interdigital transducers were created on the surface of a LiNbO3 wafer by photolithography. Leveraging the propagation characteristics of surface acoustic waves on different wetting models, the latter serve as a switch for microfluidic motion, successfully achieving selection of the driven object and demonstrating sequential reactions among multiple droplets. Also, under the excitation of standing surface acoustic waves, droplets on the wetting gradient surface move slightly in the direction of wetting gradient descent, significantly enhancing the positional accuracy of the droplets to the micrometer level. With the advantages of surface acoustic wave digital microfluidics, this technology addresses the challenges of multi-droplet digital manipulation and improved droplet positional accuracy.