Application of improved GAN-LSTM-based fake face detection technique in electronic data forensics

Due to the ease with which digital information can be altered, many digital forensic techniques have been developed to authenticate multimedia content. Similarly, a number of anti-forensic operations have recently been designed to make digital forgeries undetectable by forensic techniques. However, like the digital manipulations they are designed to hide, many anti-forensic operations leave behind their own forensically detectable traces. As a result, a digital forger must balance the trade-off between completely erasing evidence of their forgery and introducing new evidence of anti-forensic manipulation. Because a forensic investigator is typically bound by a constraint on their probability of false alarm (P_fa), they must also balance a trade-off between the accuracy with which they detect forgeries and the accuracy with which they detect the use of anti-forensics. In this paper, we analyze the interaction between a forger and a forensic investigator by examining the problem of authenticating digital videos. Specifically, we study the problem of adding or deleting a sequence of frames from a digital video. We begin by developing a theoretical model of the forensically detectable fingerprints that frame deletion or addition leaves behind, then use this model to improve upon the video frame deletion or addition detection technique proposed by Wang and Farid. Next, we propose an anti-forensic technique designed to fool video forensic techniques and develop a method for detecting the use of anti-forensics. We introduce a new set of techniques for evaluating the performance of anti-forensic operations and develop a game theoretic framework for analyzing the interplay between a forensic investigator and a forger. We use these new techniques to evaluate the performance of each of our proposed forensic and anti-forensic techniques, and identify the optimal actions of both the forger and forensic investigator.

Digital images are widely used as primary sources of evidence in today’s world, spanning security, forensics, and legal domains. However, image tampering poses a shallow technical skill barrier with the wide availability of sophisticated, easy-to-use image manipulation software. Tampered images are often used intentionally for unlawful and malicious purposes. One of the most common forms of image manipulation attack is image splicing, which is performed by combining regions from multiple source images to synthesize an artificial image that looks natural. Digital forensic measures have been widely explored in the literature to detect such type of image forgery. However, the recent growth of counter-forensics poses a threat to such forensic/security measures. Forensic techniques can be easily deceived by adopting counter-forensic manipulation of forged images. In this work, we explore different linear and non-linear filtering-based counter-forensic modifications to digital images and hence investigate the after-effects of those, in terms of severity of such manipulations in rendering state-of-the-art forensic splicing detection methods useless. In this paper, we implement two forensic image splicing detection techniques based on feature extraction from image along with machine learning and deep CNN with transfer learning. Then, different filtering techniques have been applied to the image dataset, investigating their effectiveness as a counter-forensic attack against image splicing detection. Experimental results show that the Gaussian filter and Average filter are the two most effective counter-forensic filtering methods against image splicing detection, suggesting the need for further strengthening the existing family of forensic techniques.

<p>The study identified the underlying factors that hinder the application of forensic accounting techniques in Nigeria. Data for the study were obtained from questionnaire administered on chief financial officers in Ministries, Departments and Agencies in 9 sampled States and the Federal Capital Territory, Abuja. The application of factor analysis statistical techniques collapsed the 23 measured variables into three factors. The factors were named ‘legal factor’, ‘educational factor’ and ‘political factor’ in that order of importance. It was concluded that these three factors constitute the major hindrances to the application of forensic accounting techniques for fraud prevention and detection in Nigeria. It is recommended that the legal system should be strengthened to be able to effectively handle fraud cases; educational and training institutions should embark on the training of forensic accountant; and the government should have the political will to prosecute offenders and institutionalize policies beyond what are on ground for the effective application of forensic accounting techniques in the country.</p>

In order to verify the authenticity of digital images, researchers have begun developing digital forensic techniques to identify image editing. One editing operation that has recently received increased attention is median filtering. While several median filtering detection techniques have recently been developed, their performance is degraded by JPEG compression. These techniques suffer similar degradations in performance when a small window of the image is analyzed, as is done in localized filtering or cut-and-paste detection, rather than the image as a whole. In this paper, we propose a new, robust median filtering forensic technique. It operates by analyzing the statistical properties of the median filter residual (MFR), which we define as the difference between an image in question and a median filtered version of itself. To capture the statistical properties of the MFR, we fit it to an autoregressive (AR) model. We then use the AR coefficients as features for median filter detection. We test the effectiveness of our proposed median filter detection techniques through a series of experiments. These results show that our proposed forensic technique can achieve important performance gains over existing methods, particularly at low false-positive rates, with a very small dimension of features.

Digital forensic accounting is a rapidly growing field in modern accounting, combining traditional investigative methods with cutting-edge digital technology to uncover and prevent financial fraud. One type of fraud that is a major concern in the corporate and organizational world is occupational fraud. Occupational fraud is fraudulent acts committed by employees, managers, or related insiders intended to gain personal gain at the expense of the organization. This study aims to examine the role and effectiveness of using digital forensic accounting techniques in detecting and preventing occupational fraud through a literature review of ten research articles published between 2023 and 2025. The results of the study show that developments in digital technology such as Artificial Intelligence (AI), Machine Learning (ML), Big Data Analytics, Blockchain, Digital Forensic Tools, Intelligent Automation and Deep Learning Tools have had a significant impact on improving the accuracy, speed, and efficiency of the fraud detection and investigation process. The application of these technologies enables forensic auditors to identify suspicious transaction patterns in real time, strengthen internal control systems, and improve the transparency of organizational financial data. However, the study also found challenges in the form of limited human resources with technological expertise, lack of data integration between organizational units, and data security and ethics issues. Overall, digital forensic accounting has proven to be a strategic instrument in supporting the prevention and detection of occupational fraud in the era of digital transformation.

Identifying a signal’s origin and how it was acquired is an important forensic problem. While forensic techniques currently exist to determine a signal’s acquisition history, these techniques do not account for the possibility that a signal could be compressively sensed. This is an important problem since compressive sensing techniques have seen increased popularity in recent years. In this paper, we propose a set of forensic techniques to identify signals acquired by compressive sensing. We do this by first identifying the fingerprints left in a signal by compressive sensing. We then propose two compressive sensing detection techniques that can operate on a broad class of signals. Since compressive sensing fingerprints can be confused with fingerprints left by traditional image compression techniques, we propose a forensic technique specifically designed to identify compressive sensing in digital images. In addition, we propose a technique to forensically estimate the number of compressive measurements used to acquire a signal. Through a series of experiments, we demonstrate that each of our proposed techniques can perform reliably under realistic conditions. Simulation results show that both our zero ratio detector and distribution-based detector yield perfect detections for all reasonable conditions that compressive sensing is used in applications, and the specific two-step detector for images can at least achieve probability of detection of 90% for probability of false alarm <10%. In addition, our estimator for the number of compressive measurements can well reflect the real number.

This study demonstrates various face detection and recognition techniques which have been studied till now and compares them on basis of their merits and demerits, discusses their methodologies of working and put forward a core idea of how face detection is done, mentioning about very basic term, so that any person who is not that good in technology can understand it and dive deep into this field. Some of the face detection techniques we will look into will be geometric-based face detection, feature-based face detection, and Haar-like feature based-face detection, giving special emphasis on Haar-like features-based face detection. In face recognition techniques, we looked into the lazy learner’s face recognition approach, neural networks face recognition approach, and holistic face recognition approaches. Objective of the study is also to demonstrate about preparing a model which detects faces in a real-time environment. Face detection, nowadays, is the most primary check in any security system. So, automation in detecting faces will prove helpful. This model, rather than any other model, works on real-time data provided. Our model works on the fundamentals of the K-nearest neighbors algorithm, Haar cascade classifier (an object detection technique), and OpenCV (an open-source python library for computer vision, machine learning, and image processing).KeywordsK-nearest neighborPrincipal component analysisOpen-source computer visionArtificial neural networkRectified linear activation unitAutomated teller machine

A technique for face detection in the image is proposed, which is based on binarization, scaling, and segmentation of the image, followed by the determination of the largest connected component that matches the image of the face.\n\nModern methods of binarization, scaling, and taxonomic image segmentation have one or more of the following disadvantages: they have a high computational complexity; require the determination of parameter values. Taxonomic image segmentation methods may have additional disadvantages: they do not allow noise and outliers selection; clusters can’t have different shapes and sizes, and their number is fixed.\n\nDue to this, to improve the efficiency of face detection techniques, the methods of binarization, scaling and taxonomic segmentation needs to be improved.\n\nA binarization method is proposed, the distinction of which is the use of the image background. This allows to simplify the process of scaling and segmentation (since all the pixels in the background are represented by the same color), non-uniform brightness of the face, and not to use the threshold settings and additional parameters.\n\nA binary image scaling method is proposed, the distinction of which is the use of an arithmetic mean filter with threshold processing and fast wavelet transform. This allows to speed up the image segmentation process by about P2 times, where P is the scaling parameter, and not to use the time-consuming procedure for determining.\n\nA binary scaled image segmentation method is proposed, the distinction of which is the use of density clustering. This allows to separate areas of the face of non-uniform brightness from the image background, noise and outliers. It also allows clusters to have different shapes and sizes, to not require setting the number of clusters and additional parameters.\n\nTo determine the scaling parameter, numerous studies were conducted in this work, which concluded that the dependence of the segmentation time on the scaling parameter is close to exponential. It was also found that for small P, where P is the scaling parameter, the quality of face detection deteriorates slightly.\nThe proposed technique for face detection in image based on binarization, scaling and segmentation can be used in intelligent computer systems for biometric identification of a person by the face image

Web application attacks are an increasingly important area in information security and digital forensics. It has been observed that attackers are developing the capability to bypass security controls and launch a large number of sophisticated attacks. Several attempts have been made to address these attacks using a wide range of technology and one of the greatest challenges is responding to new and unknown attacks in an effective way. This study aims to investigate the techniques and solutions used to detect attacks, such as firewalls, intrusion detection systems, honeypots and forensic techniques. Data mining and machine learning techniques, which attempt to address traditional technology shortcomings and produce more effective solutions, are also investigated. It was aimed to contribute to this growing area of research by exploring more intelligent and convenient techniques for web application attack detection by focusing on the data mining techniques in forensics.

Digital forensics is a vital part of almost every criminal investigation given the amount of information available and the opportunities offered by electronic data to investigate and evidence a crime. However, in criminal justice proceedings, these electronic pieces of evidence are often considered with the utmost suspicion and uncertainty, although, on occasions are justifiable. Presently, the use of scientifically unproven forensic techniques are highly criticized in legal proceedings. Nevertheless, the exceedingly distinct and dynamic characteristics of electronic data, in addition to the current legislation and privacy laws remain as challenging aspects for systematically attesting evidence in a court of law. This article presents a comprehensive study to examine the issues that are considered essential to discuss and resolve, for the proper acceptance of evidence based on scientific grounds. Moreover, the article explains the state of forensics in emerging sub-fields of digital technology such as, cloud computing, social media, and the Internet of Things (IoT), and reviewing the challenges which may complicate the process of systematic validation of electronic evidence. The study further explores various solutions previously proposed, by researchers and academics, regarding their appropriateness based on their experimental evaluation. Additionally, this article suggests open research areas, highlighting many of the issues and problems associated with the empirical evaluation of these solutions for immediate attention by researchers and practitioners. Notably, academics must react to these challenges with appropriate emphasis on methodical verification. Therefore, for this purpose, the issues in the experiential validation of practices currently available are reviewed in this study. The review also discusses the struggle involved in demonstrating the reliability and validity of these approaches with contemporary evaluation methods. Furthermore, the development of best practices, reliable tools and the formulation of formal testing methods for digital forensic techniques are highlighted which could be extremely useful and of immense value to improve the trustworthiness of electronic evidence in legal proceedings.

Fraud is a global problem. However, the rate and nature of fraud in Nigeria is quite alarming. The study therefore examined the behavioural intention of accounting practitioner to use forensic accounting techniques in fraud prevention and detection in Nigeria. Data were collected from from 9 states and the Federal Capital Territory, Abuja. The Structural equation modelling was adopted in analyzing the data. Seven propositions were tested in the study. From the results, all the propositions were supported. It was concluded from this study that accountants will accept use of forensic accounting in the prevention and detection of fraud if they understand the benefits, risk, fraud susceptibility and fraud severity in their establishment. It was recommended that educational activities of training institution should therefore be directed toward increasing awareness on forensic accounting.

Malware that exploits user privacy has increased in recent decades, and this trend has been linked to shifting international regulations, the expansion of Internet services, and the growth of electronic commerce. Furthermore, it is very challenging to detect privacy malware that uses obfuscation as an evasion tactic due to its behaviour, resilience, and adaptability during runtime. Forensic techniques, such as memory dumping analysis, must be used to enable a system to identify and classify patterns and behaviours that facilitate its eventual identification. This research developed a deep learning model for malware classification on an obfuscated malware dataset, called the MalwareMemoryDump dataset. It implemented transfer learning (TL) to adapt the trained model to NF-TON-IoT and UNSW-NB15, improving intrusion detection in IoT and network traffic. We conducted extensive experiments showing improved accuracy and efficiency in cross-domain detection scenarios. Further, we demonstrate that transfer learning minimises training time and computational requirements compared to training separate models from scratch. Additionally, it offers XAI-based explainability to enhance model transparency and interoperability. We demonstrated the effectiveness of the proposed model in handling diverse heterogeneous cybersecurity threats across memory-based malware analysis, IoT security, and traditional network intrusion detection. The effectiveness of the proposed methodology is evaluated using several key metrics to demonstrate its advantages over conventional methods. Experimental findings show that the proposed framework attains 99.9% accuracy on the MalwareMemoryDump dataset, 96% on the NF-Ton-IoT dataset and UNSW-NB15 datasets. Because of its innovative methodology and ability to generalise datasets, the model is a highly effective approach that outperforms many of the most recent malware detection and other security techniques.

Face detection is a challenging and important problem in Computer Vision. In most of the face recognition systems, face detection is used in order to locate the faces in the images. There are different methods for detecting faces in images. One of these methods is to try to find faces in the part of the image that contains human skin. This can be done by using the information of human skin color. Skin detection can be challenging due to factors such as the differences in illumination, different cameras, ranges of skin colors due to different ethnicities, and other variations. Neural networks have been used for detecting human skin. Different methods have been applied to neural networks in order to increase the detection rate of the human skin. The resulting image is then used in the detection phase. The resulting image consists of several components and in the face detection phase, the faces are found by just searching those components. If the components consist of just faces, then the faces can be detected using correlation. Eye and lip detections have also been investigated using different methods, using information from different color spaces. The speed of face detection methods using color images is compared with other face detection methods.

Human face detection plays considerably important role in various biometric applications like crowd surveillance, photography, human-computer interaction, tracking, automatic target recognition, artificial intelligence and various security applications. Varying illumination conditions, color variance, brightness, pose variations are major challenging problems for facial detection. Skin color based segmentation and AdaBoost based facial detection scheme are the two most widely used techniques for face detection. But skin color segmentation method has very high false positive detection rate in images with complicated background and AdaBoost algorithm does not provide desired results for detecting images having multiple pose and multiple faces. Apart from this, AdaBoost approach has higher accuracy, but slower speed and skin color segmentation method has a faster speed of detection, but lower accuracy; and. So our paper proposes a novel facial detection scheme based on the integration of YCgCr based skin color segmentation and improved AdaBoost algorithm. Also morphological operators are applied to improve the detection performance. From the experimental results, it can be deduced that the proposed face detection algorithm improves the detection speed, accuracy and capable of real time face detection. Simulation results are used to show that our proposed method achieves accuracy of approximately 97% and has considerably good performance on images having complex background and can detect faces of various sizes, postures and expressions, under uncontrolled lighting environments.

In this paper we propose a novel face detection method based on the vanishing point of vertical lines in the scene to improve system performance in a common surveillance application. While most existing face datasets and detection techniques are based on the assumption that the camera has a similar height as the target faces, in practical situations the camera may be installed at different heights. Such discrepancy often degrades the detection performance of algorithms based on learning with certain (e.g., frontal) face orientation. In this paper we propose a transformation to rectify face images (video frames) such that it is not necessary to collect training data of different face orientations. Furthermore, with the proposed method there is no need to perform complex camera calibration. The only required information is the vanishing point of vertical lines, which can often be estimated easily. Experiments show prominent improvements in face detection performance can be obtained with the proposed image transformation.