Network intrusion forensics is an important extension to present security infrastructure,and is becoming the focus of forensics research field.However,comparison with sophisticated multi-stage attacks and volume of se...Network intrusion forensics is an important extension to present security infrastructure,and is becoming the focus of forensics research field.However,comparison with sophisticated multi-stage attacks and volume of sensor data,current practices in network forensic analysis are to manually examine,an error prone,labor-intensive and time consuming process.To solve these problems,in this paper we propose a digital evidence fusion method for network forensics with Dempster-Shafer theory that can detect efficiently computer crime in networked environments,and fuse digital evidence from different sources such as hosts and sub-networks automatically.In the end,we evaluate the method on well-known KDD Cup1999 dataset.The results prove our method is very effective for real-time network forensics,and can provide comprehensible messages for a forensic investigators.展开更多
With the rapid advancement of visual generative models such as Generative Adversarial Networks(GANs)and stable Diffusion,the creation of highly realistic Deepfake through automated forgery has significantly progressed...With the rapid advancement of visual generative models such as Generative Adversarial Networks(GANs)and stable Diffusion,the creation of highly realistic Deepfake through automated forgery has significantly progressed.This paper examines the advancements inDeepfake detection and defense technologies,emphasizing the shift from passive detection methods to proactive digital watermarking techniques.Passive detection methods,which involve extracting features from images or videos to identify forgeries,encounter challenges such as poor performance against unknown manipulation techniques and susceptibility to counter-forensic tactics.In contrast,proactive digital watermarking techniques embed specificmarkers into images or videos,facilitating real-time detection and traceability,thereby providing a preemptive defense againstDeepfake content.We offer a comprehensive analysis of digitalwatermarking-based forensic techniques,discussing their advantages over passivemethods and highlighting four key benefits:real-time detection,embedded defense,resistance to tampering,and provision of legal evidence.Additionally,the paper identifies gaps in the literature concerning proactive forensic techniques and suggests future research directions,including cross-domain watermarking and adaptive watermarking strategies.By systematically classifying and comparing existing techniques,this review aims to contribute valuable insights for the development of more effective proactive defense strategies in Deepfake forensics.展开更多
The smart home platform integrates with Internet of Things(IoT)devices,smartphones,and cloud servers,enabling seamless and convenient services.It gathers and manages extensive user data,including personal information,...The smart home platform integrates with Internet of Things(IoT)devices,smartphones,and cloud servers,enabling seamless and convenient services.It gathers and manages extensive user data,including personal information,device operations,and patterns of user behavior.Such data plays an essential role in criminal inves-tigations,highlighting the growing importance of specialized smart home forensics.Given the rapid advancement in smart home software and hardware technologies,many companies are introducing new devices and services that expand the market.Consequently,scalable and platform-specific forensic research is necessary to support efficient digital investigations across diverse smart home ecosystems.This study thoroughly examines the core components and structures of smart homes,proposing a generalized architecture that represents various operational environments.A three-stage smart home forensics framework is introduced:(1)analyzing application functions to infer relevant data,(2)extracting and processing data from interconnected devices,and(3)identifying data valuable for investigative purposes.The framework’s applicability is validated using testbeds from Samsung SmartThings and Xiaomi Mi Home platforms,offering practical insights for real-world forensic applications.The results demonstrate that the proposed forensic framework effectively acquires and classifies relevant digital evidence in smart home platforms,confirming its practical applicability in smart home forensic investigations.展开更多
Images and videos play an increasingly vital role in daily life and are widely utilized as key evidentiary sources in judicial investigations and forensic analysis.Simultaneously,advancements in image and video proces...Images and videos play an increasingly vital role in daily life and are widely utilized as key evidentiary sources in judicial investigations and forensic analysis.Simultaneously,advancements in image and video processing technologies have facilitated the widespread availability of powerful editing tools,such as Deepfakes,enabling anyone to easily create manipulated or fake visual content,which poses an enormous threat to social security and public trust.To verify the authenticity and integrity of images and videos,numerous approaches have been proposed,which are primarily based on content analysis and their effectiveness is susceptible to interference from various image or video post-processing operations.Recent research has highlighted the potential of file containers analysis as a promising forensic approach that offers efficient and interpretable results.However,there is still a lack of review articles on this kind of approach.In order to fill this gap,we present a comprehensive review of file containers-based image and video forensics in this paper.Specifically,we categorize the existing methods into two distinct stages,qualitative analysis and quantitative analysis.In addition,an overall framework is proposed to organize the exiting approaches.Then,the advantages and disadvantages of the schemes used across different forensic tasks are provided.Finally,we outline the trends in this research area,aiming to provide valuable insights and technical guidance for future research.展开更多
Artifcial Intelligence (AI) is being applied to improve the efciency of software systems used in various domains, especially in the health and forensic sciences. Explainable AI (XAI) is one of the felds of AI that int...Artifcial Intelligence (AI) is being applied to improve the efciency of software systems used in various domains, especially in the health and forensic sciences. Explainable AI (XAI) is one of the felds of AI that interprets and explains the methods used in AI. One of the techniques used in XAI to provide such interpretations is by computing the rel-evanceof the input features to the output of an AI model. File fragment classifcation is one of the vital issues of fle carving in Cyber Forensics (CF) and becomes challenging when the flesystem metadata is missing. Other major challenges it faces are: proliferation of fle formats, fle embeddings, automation, We leverage and utilize interpretations provided by XAI to optimize the classifcation of fle fragments and propose a novel sifting approach, named SIFT (Sifting File Types). SIFT employs TF-IDF to assign weight to a byte (feature), which is used to select features from a fle fragment. Threshold-based LIME and SHAP (the two XAI techniques) feature relevance values are computed for the selected features to optimize fle fragment classifcation. To improve multinomial classifcation, a Multilayer Per-ceptronmodel is developed and optimized with fve hidden layers, each layer with i × n neurons, where i = the layer number and n = the total number of classes in the dataset. When tested with 47,482 samples of 20 fle types (classes), SIFT achieves a detection rate of 82.1% and outperforms the other state-of-the-art techniques by at least 10%. To the best of our knowledge, this is the frst efort of applying XAI in CF for optimizing fle fragment classifcation.展开更多
Electric Vehicle Charging Systems(EVCS)are increasingly vulnerable to cybersecurity threats as they integrate deeply into smart grids and Internet ofThings(IoT)environments,raising significant security challenges.Most...Electric Vehicle Charging Systems(EVCS)are increasingly vulnerable to cybersecurity threats as they integrate deeply into smart grids and Internet ofThings(IoT)environments,raising significant security challenges.Most existing research primarily emphasizes network-level anomaly detection,leaving critical vulnerabilities at the host level underexplored.This study introduces a novel forensic analysis framework leveraging host-level data,including system logs,kernel events,and Hardware Performance Counters(HPC),to detect and analyze sophisticated cyberattacks such as cryptojacking,Denial-of-Service(DoS),and reconnaissance activities targeting EVCS.Using comprehensive forensic analysis and machine learning models,the proposed framework significantly outperforms existing methods,achieving an accuracy of 98.81%.The findings offer insights into distinct behavioral signatures associated with specific cyber threats,enabling improved cybersecurity strategies and actionable recommendations for robust EVCS infrastructure protection.展开更多
Objective Current autosomal short tandem repeat(STR)assays can analyze the zygotic composition by comparing the allelic genes at each locus of complete hydatidiform moles(CHM),with a maternal genotype serving as an es...Objective Current autosomal short tandem repeat(STR)assays can analyze the zygotic composition by comparing the allelic genes at each locus of complete hydatidiform moles(CHM),with a maternal genotype serving as an essential reference for comparative analysis.However,their application in pathology represents a challenge because of deficiency or contamination of maternal-origin tissues.This study aimed to develop a novel STR genotyping method for identifying CHM genotypes without a maternal component.Methods Samples with the pathologic description of molar pregnancy were collected.Routine hematoxylin–eosin(HE)staining and p57 immunohistochemistry staining were conducted in accordance with standard guidelines.A novel 26-plex system was explored to classify CHM and diploid pregnancies.The system combined 22 STRs on chromosomes 21/18/13/X,3 sex loci,and 1 quality control marker(TAF9L),enabling molecular diagnosis in the absence of maternal tissue.At last,traditional DNA typing based on villi and decidua(maternal component)of each case was used for result consistency analysis.Results CHM and nonmolar abortus could not be distinguished by the basic HE staining with no fetal evidence or other prominent features.DNA typing was successfully processed for all cases according to the novel 26-plex and traditional system.CHM(46XX)diagnosis required single A-STR/X-STR peaks and absent Y-chromosome markers,excluding chromosomal abnormalities via TAF9L analysis.When the villous tissue analysis revealed single peaks at X-STR/SRY loci,a 1:1 amelogenin ratio,and a 2:1 TAF9L peak ratio,these results overlapped with those of 46XY hydropic abortus or CHM.Notably,p57 immunohistochemical staining resolved the ambiguity.Consistency with traditional DNA genotyping confirmed system accuracy.This multiplex assay enhanced reliability in mole diagnosis,supporting clinical differentiation and genetic counseling.Conclusion This study presents a rapid and cost-effective assay for the genotypic identification of CHM without the need for a maternal component.The method combined the characteristics of STR loci distributed across different chromosomes and developed the clinic application of forensic biomarkers.展开更多
In today’s digital era,the rapid evolution of image editing technologies has brought about a significant simplification of image manipulation.Unfortunately,this progress has also given rise to the misuse of manipulat...In today’s digital era,the rapid evolution of image editing technologies has brought about a significant simplification of image manipulation.Unfortunately,this progress has also given rise to the misuse of manipulated images across various domains.One of the pressing challenges stemming from this advancement is the increasing difficulty in discerning between unaltered and manipulated images.This paper offers a comprehensive survey of existing methodologies for detecting image tampering,shedding light on the diverse approaches employed in the field of contemporary image forensics.The methods used to identify image forgery can be broadly classified into two primary categories:classical machine learning techniques,heavily reliant on manually crafted features,and deep learning methods.Additionally,this paper explores recent developments in image forensics,placing particular emphasis on the detection of counterfeit colorization.Image colorization involves predicting colors for grayscale images,thereby enhancing their visual appeal.The advancements in colorization techniques have reached a level where distinguishing between authentic and forged images with the naked eye has become an exceptionally challenging task.This paper serves as an in-depth exploration of the intricacies of image forensics in the modern age,with a specific focus on the detection of colorization forgery,presenting a comprehensive overview of methodologies in this critical field.展开更多
Advances in technological developments in Geographic Information Systems(GIS)has enabled the application of GIS in landscape mapping,environmental management,natural hazard risk and disaster management.As geographical...Advances in technological developments in Geographic Information Systems(GIS)has enabled the application of GIS in landscape mapping,environmental management,natural hazard risk and disaster management.As geographical information becomes more widely available through satellite and aerial imagery,the cost of software decreases and GIS expertise expands,it is most likely that the use of GIS will increase.The methodology has practical applications for police,crime scene investigators and forensic geoscientists.The aim is to develop GIS use in forensic search beyond mapping to offer a set of decision support tools that utilise the spatial analytical capabilities of GIS.This enables better management and understanding of the complicated and interrelated nature of a ground search.展开更多
In this age when most organizations make use of cloud computing,it is important to not only protect cloud computing resources from cyber⁃attacks but also investigate these attacks.During forensic investigations in a c...In this age when most organizations make use of cloud computing,it is important to not only protect cloud computing resources from cyber⁃attacks but also investigate these attacks.During forensic investigations in a cloud environment,the investigators fall on service providers for pieces of evidence like log files.The challenge,however,is the integrity of these logs provided by the service providers.To this end,we propose a blockchain⁃based log verification system called BlogVerifier that uses a decentralized approach to solve forensics issues in the cloud.BlogVerifier extracts logs produced in cloud environments,hashes these logs and stores the hashed values as transactional values on the blockchain.The transactions are then merged into blocks and shared on the blockchain.The proposed system also ensures the continuation of an investigation even when the primary source of a log is compromised by using encryption and smart contracts.The proposed system also makes it possible for any stakeholder involved in the forensic process to verify the authenticity of log files.The performance results show that BlogVerifier can be integrated into the cloud environment without any significant impact on system resources and increase in computational cost.展开更多
The proliferation of cloud computing and internet of things has led to the connectivity of states and nations(developed and developing countries)worldwide in which global network provide platform for the connection.Di...The proliferation of cloud computing and internet of things has led to the connectivity of states and nations(developed and developing countries)worldwide in which global network provide platform for the connection.Digital forensics is a field of computer security that uses software applications and standard guidelines which support the extraction of evidences from any computer appliances which is perfectly enough for the court of law to use and make a judgment based on the comprehensiveness,authenticity and objectivity of the information obtained.Cybersecurity is of major concerned to the internet users worldwide due to the recent form of attacks,threat,viruses,intrusion among others going on every day among internet of things.However,it is noted that cybersecurity is based on confidentiality,integrity and validity of data.The aim of this work is make a systematic review on the application of machine learning algorithms to cybersecurity and cyber forensics and pave away for further research directions on the application of deep learning,computational intelligence,soft computing to cybersecurity and cyber forensics.展开更多
Digital forensics aims to uncover evidence of cybercrimes within compromised systems.These cybercrimes are often perpetrated through the deployment of malware,which inevitably leaves discernible traces within the comp...Digital forensics aims to uncover evidence of cybercrimes within compromised systems.These cybercrimes are often perpetrated through the deployment of malware,which inevitably leaves discernible traces within the compromised systems.Forensic analysts are tasked with extracting and subsequently analyzing data,termed as artifacts,from these systems to gather evidence.Therefore,forensic analysts must sift through extensive datasets to isolate pertinent evidence.However,manually identifying suspicious traces among numerous artifacts is time-consuming and labor-intensive.Previous studies addressed such inefficiencies by integrating artificial intelligence(AI)technologies into digital forensics.Despite the efforts in previous studies,artifacts were analyzed without considering the nature of the data within them and failed to prove their efficiency through specific evaluations.In this study,we propose a system to prioritize suspicious artifacts from compromised systems infected with malware to facilitate efficient digital forensics.Our system introduces a double-checking method that recognizes the nature of data within target artifacts and employs algorithms ideal for anomaly detection.The key ideas of this method are:(1)prioritize suspicious artifacts and filter remaining artifacts using autoencoder and(2)further prioritize suspicious artifacts and filter remaining artifacts using logarithmic entropy.Our evaluation demonstrates that our system can identify malicious artifacts with high accuracy and that its double-checking method is more efficient than alternative approaches.Our system can significantly reduce the time required for forensic analysis and serve as a reference for future studies.展开更多
Detecting hate speech automatically in social media forensics has emerged as a highly challenging task due tothe complex nature of language used in such platforms. Currently, several methods exist for classifying hate...Detecting hate speech automatically in social media forensics has emerged as a highly challenging task due tothe complex nature of language used in such platforms. Currently, several methods exist for classifying hatespeech, but they still suffer from ambiguity when differentiating between hateful and offensive content and theyalso lack accuracy. The work suggested in this paper uses a combination of the Whale Optimization Algorithm(WOA) and Particle Swarm Optimization (PSO) to adjust the weights of two Multi-Layer Perceptron (MLPs)for neutrosophic sets classification. During the training process of the MLP, the WOA is employed to exploreand determine the optimal set of weights. The PSO algorithm adjusts the weights to optimize the performanceof the MLP as fine-tuning. Additionally, in this approach, two separate MLP models are employed. One MLPis dedicated to predicting degrees of truth membership, while the other MLP focuses on predicting degrees offalse membership. The difference between these memberships quantifies uncertainty, indicating the degree ofindeterminacy in predictions. The experimental results indicate the superior performance of our model comparedto previous work when evaluated on the Davidson dataset.展开更多
The AGCU X Plus STR system is a newly developed multiplex PCR kit that detects 32 X-chromosomal STR loci simultaneously.These are DXS6807,DXS9895,linkage group 1(DXS10148,DXS10135,DXS8378),DXS9902,DXS6795,DXS6810,DXS1...The AGCU X Plus STR system is a newly developed multiplex PCR kit that detects 32 X-chromosomal STR loci simultaneously.These are DXS6807,DXS9895,linkage group 1(DXS10148,DXS10135,DXS8378),DXS9902,DXS6795,DXS6810,DXS10159,DXS10162,DXS10164,DXS7132,linkage group 2(DXS10079,DXS10074,DXS10075),DXS981,DXS6800,DXS6803,DXS6809,DXS6789,DXS7424,DXS101,DXS7133,GATA172D05,GATA165B12,linkage group 3(DXS10103,HPRTB,DXS10101),GATA31E08 and linkage group 4(DXS8377,DXS10134,DXS7423).A major advantage of this kit is that it takes into account linkage between loci,in addition to detecting more X-STR loci.In order to evaluate the forensic application of 32 X-STR fl uorescence amplifi cation system,PCR settings,sensitivity,species specifi city,stability,DNA mixtures,concordance,stutter,sizing precision,and population genetics investigation were evaluated according to the Scientific Working Group on DNA Analysis Methods(SWGDAM)developmental validation guidelines.The study showed that the genotyping results of each locus were signifi cantly accurate when the DNA template was at least 62.5 pg.Complete profi les were obtained for the 1∶1 and 1∶3 combinations.A total of 209 unrelated individuals from Southern Chinese Han community,consisting of 84 females and 125 males,were selected for population studies,and 285 allele profi les were detected from 32 X-STR loci.The polymorphism information content(PIC)ranged from 0.2721 in DXS6800,to 0.9105 in DXS10135,with an average of 0.6798.DXS10135(PIC=0.9105)was the most polymorphic locus,with discrimination power(DP)of 0.9164 and 0.9871 for the male and female.The cumulative PD_(F),PD_(M),MEC_(trio) and MEC_(duo) valu es were all greater than 0.999999999.There were 78 different DXS10103-HPRTB-DXS10101 haplotypes among the 125 males,and the haplotype diversity was 0.9810.There was no signifi cant difference in the cumulative PD_(F),PD_(M),MEC_(trio) and MEC_(duo) values whether considering linkage or not.In summary,the new X-STR multiplex typing system is effective and reliable,which can be useful in human genetic analysis and kinship testing as a potent complement to autosomal STR typing.展开更多
Signature verification involves vague situations in which a signature could resemble many reference samples ormight differ because of handwriting variances. By presenting the features and similarity score of signature...Signature verification involves vague situations in which a signature could resemble many reference samples ormight differ because of handwriting variances. By presenting the features and similarity score of signatures from thematching algorithm as fuzzy sets and capturing the degrees of membership, non-membership, and indeterminacy,a neutrosophic engine can significantly contribute to signature verification by addressing the inherent uncertaintiesand ambiguities present in signatures. But type-1 neutrosophic logic gives these membership functions fixed values,which could not adequately capture the various degrees of uncertainty in the characteristics of signatures. Type-1neutrosophic representation is also unable to adjust to various degrees of uncertainty. The proposed work exploresthe type-2 neutrosophic logic to enable additional flexibility and granularity in handling ambiguity, indeterminacy,and uncertainty, hence improving the accuracy of signature verification systems. Because type-2 neutrosophiclogic allows the assessment of many sources of ambiguity and conflicting information, decision-making is moreflexible. These experimental results show the possible benefits of using a type-2 neutrosophic engine for signatureverification by demonstrating its superior handling of uncertainty and variability over type-1, which eventuallyresults in more accurate False Rejection Rate (FRR) and False Acceptance Rate (FAR) verification results. In acomparison analysis using a benchmark dataset of handwritten signatures, the type-2 neutrosophic similaritymeasure yields a better accuracy rate of 98% than the type-1 95%.展开更多
The development of high technology,for public life to provide a justification at the same time,also encouraged the spirit of cybercrime,to become more and more rampant.In network crime,electronic data is usually used ...The development of high technology,for public life to provide a justification at the same time,also encouraged the spirit of cybercrime,to become more and more rampant.In network crime,electronic data is usually used as the main evidence to determine the facts of the crime and plays an important role in the smooth trial of the case.But because electronic data on dependent,concealment,easy destructive strong science and technology,the forensics work is now in trouble.The mature use of blockchain technology can avoid existing problems to a certain extent,which is helpful to the smooth progress of electronic forensics.This paper on electronic evidence how to more effectively,combined with research blockchain technology,improve the efficiency of electronic evidence collection work.展开更多
In recent years,visual facial forgery has reached a level of sophistication that humans cannot identify fraud,which poses a significant threat to information security.A wide range of malicious applications have emerge...In recent years,visual facial forgery has reached a level of sophistication that humans cannot identify fraud,which poses a significant threat to information security.A wide range of malicious applications have emerged,such as deepfake,fake news,defamation or blackmailing of celebrities,impersonation of politicians in political warfare,and the spreading of rumours to attract views.As a result,a rich body of visual forensic techniques has been proposed in an attempt to stop this dangerous trend.However,there is no comprehensive,fair,and unified performance evaluation to enlighten the community on best performing methods.The authors present a systematic benchmark beyond traditional surveys that provides in-depth insights into facial forgery and facial forensics,grounding on robustness tests such as contrast,brightness,noise,resolution,missing information,and compression.The authors also provide a practical guideline of the benchmarking results,to determine the characteristics of the methods that serve as a comparative reference in this never-ending war between measures and countermeasures.The authors’source code is open to the public.展开更多
Vehicle data is one of the important sources of traffic accident digital forensics.We propose a novel method using long short-term memory-deep belief network by binary encoding(LSTM-BiDBN)controller area network ident...Vehicle data is one of the important sources of traffic accident digital forensics.We propose a novel method using long short-term memory-deep belief network by binary encoding(LSTM-BiDBN)controller area network identifier(CAN ID)to extract the event sequence of CAN IDs and the semantic of CAN IDs themselves.Instead of detecting attacks only aimed at a specific CAN ID,the proposed method fully considers the potential interaction between electronic control units.By this means,we can detect whether the vehicle has been invaded by the outside,to online determine the responsible party of the accident.We use our LSTM-BiDBN to distinguish attack-free and abnormal situations on CAN-intrusion-dataset.Experimental results show that our proposed method is more effective in identifying anomalies caused by denial of service attack,fuzzy attack and impersonation attack with an accuracy value of 97.02%,a false-positive rate of 6.09%,and a false-negative rate of 1.94%compared with traditional methods.展开更多
In the field of image forensics,image tampering detection is a critical and challenging task.Traditional methods based on manually designed feature extraction typically focus on a specific type of tampering operation,...In the field of image forensics,image tampering detection is a critical and challenging task.Traditional methods based on manually designed feature extraction typically focus on a specific type of tampering operation,which limits their effectiveness in complex scenarios involving multiple forms of tampering.Although deep learningbasedmethods offer the advantage of automatic feature learning,current approaches still require further improvements in terms of detection accuracy and computational efficiency.To address these challenges,this study applies the UNet 3+model to image tampering detection and proposes a hybrid framework,referred to as DDT-Net(Deep Detail Tracking Network),which integrates deep learning with traditional detection techniques.In contrast to traditional additive methods,this approach innovatively applies amultiplicative fusion technique during downsampling,effectively combining the deep learning feature maps at each layer with those generated by the Bayar noise stream.This design enables noise residual features to guide the learning of semantic features more precisely and efficiently,thus facilitating comprehensive feature-level interaction.Furthermore,by leveraging the complementary strengths of deep networks in capturing large-scale semantic manipulations and traditional algorithms’proficiency in detecting fine-grained local traces,the method significantly enhances the accuracy and robustness of tampered region detection.Compared with other approaches,the proposed method achieves an F1 score improvement exceeding 30% on the DEFACTO and DIS25k datasets.In addition,it has been extensively validated on other datasets,including CASIA and DIS25k.Experimental results demonstrate that this method achieves outstanding performance across various types of image tampering detection tasks.展开更多
Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection m...Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection mechanisms is paramount.This research introduces a novel real-time deepfake video detection framework by analyzing gaze and blink patterns,addressing the spatial-temporal challenges unique to gaze and blink anomalies using the TimeSformer and hybrid Transformer-CNN models.The TimeSformer architecture leverages spatial-temporal attention mechanisms to capture fine-grained blinking intervals and gaze direction anomalies.Compared to state-of-the-art traditional convolutional models like MesoNet and EfficientNet,which primarily focus on global facial features,our approach emphasizes localized eye-region analysis,significantly enhancing detection accuracy.We evaluate our framework on four standard datasets:FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb.The proposed framework results reveal higher accuracy,with the TimeSformer model achieving accuracies of 97.5%,96.3%,95.8%,and 97.1%,and with the hybrid Transformer-CNN model demonstrating accuracies of 92.8%,91.5%,90.9%,and 93.2%,on FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb datasets,respectively,showing robustness in distinguishing manipulated from authentic videos.Our research provides a robust state-of-the-art framework for real-time deepfake video detection.This novel study significantly contributes to video forensics,presenting scalable and accurate real-world application solutions.展开更多
基金supported by the National Natural Science Foundation of China under Grant No.60903166 the National High Technology Research and Development Program of China（863 Program） under Grants No.2012AA012506,No.2012AA012901,No.2012AA012903+9 种基金 Specialized Research Fund for the Doctoral Program of Higher Education of China under Grant No.20121103120032 the Humanity and Social Science Youth Foundation of Ministry of Education of China under Grant No.13YJCZH065 the Opening Project of Key Lab of Information Network Security of Ministry of Public Security（The Third Research Institute of Ministry of Public Security） under Grant No.C13613 the China Postdoctoral Science Foundation General Program of Science and Technology Development Project of Beijing Municipal Education Commission of China under Grant No.km201410005012 the Research on Education and Teaching of Beijing University of Technology under Grant No.ER2013C24 the Beijing Municipal Natural Science Foundation Sponsored by Hunan Postdoctoral Scientific Program Open Research Fund of Beijing Key Laboratory of Trusted Computing Funds for the Central Universities, Contract No.2012JBM030
文摘Network intrusion forensics is an important extension to present security infrastructure,and is becoming the focus of forensics research field.However,comparison with sophisticated multi-stage attacks and volume of sensor data,current practices in network forensic analysis are to manually examine,an error prone,labor-intensive and time consuming process.To solve these problems,in this paper we propose a digital evidence fusion method for network forensics with Dempster-Shafer theory that can detect efficiently computer crime in networked environments,and fuse digital evidence from different sources such as hosts and sub-networks automatically.In the end,we evaluate the method on well-known KDD Cup1999 dataset.The results prove our method is very effective for real-time network forensics,and can provide comprehensible messages for a forensic investigators.
基金supported by the National Fund Cultivation Project from China People’s Police University(Grant Number:JJPY202402)National Natural Science Foundation of China(Grant Number:62172165).
文摘With the rapid advancement of visual generative models such as Generative Adversarial Networks(GANs)and stable Diffusion,the creation of highly realistic Deepfake through automated forgery has significantly progressed.This paper examines the advancements inDeepfake detection and defense technologies,emphasizing the shift from passive detection methods to proactive digital watermarking techniques.Passive detection methods,which involve extracting features from images or videos to identify forgeries,encounter challenges such as poor performance against unknown manipulation techniques and susceptibility to counter-forensic tactics.In contrast,proactive digital watermarking techniques embed specificmarkers into images or videos,facilitating real-time detection and traceability,thereby providing a preemptive defense againstDeepfake content.We offer a comprehensive analysis of digitalwatermarking-based forensic techniques,discussing their advantages over passivemethods and highlighting four key benefits:real-time detection,embedded defense,resistance to tampering,and provision of legal evidence.Additionally,the paper identifies gaps in the literature concerning proactive forensic techniques and suggests future research directions,including cross-domain watermarking and adaptive watermarking strategies.By systematically classifying and comparing existing techniques,this review aims to contribute valuable insights for the development of more effective proactive defense strategies in Deepfake forensics.
文摘The smart home platform integrates with Internet of Things(IoT)devices,smartphones,and cloud servers,enabling seamless and convenient services.It gathers and manages extensive user data,including personal information,device operations,and patterns of user behavior.Such data plays an essential role in criminal inves-tigations,highlighting the growing importance of specialized smart home forensics.Given the rapid advancement in smart home software and hardware technologies,many companies are introducing new devices and services that expand the market.Consequently,scalable and platform-specific forensic research is necessary to support efficient digital investigations across diverse smart home ecosystems.This study thoroughly examines the core components and structures of smart homes,proposing a generalized architecture that represents various operational environments.A three-stage smart home forensics framework is introduced:(1)analyzing application functions to infer relevant data,(2)extracting and processing data from interconnected devices,and(3)identifying data valuable for investigative purposes.The framework’s applicability is validated using testbeds from Samsung SmartThings and Xiaomi Mi Home platforms,offering practical insights for real-world forensic applications.The results demonstrate that the proposed forensic framework effectively acquires and classifies relevant digital evidence in smart home platforms,confirming its practical applicability in smart home forensic investigations.
基金supported in part by Natural Science Foundation of Hubei Province of China under Grant 2023AFB016the 2022 Opening Fund for Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering under Grant 2022SDSJ02the Construction Fund for Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering under Grant 2019ZYYD007.
文摘Images and videos play an increasingly vital role in daily life and are widely utilized as key evidentiary sources in judicial investigations and forensic analysis.Simultaneously,advancements in image and video processing technologies have facilitated the widespread availability of powerful editing tools,such as Deepfakes,enabling anyone to easily create manipulated or fake visual content,which poses an enormous threat to social security and public trust.To verify the authenticity and integrity of images and videos,numerous approaches have been proposed,which are primarily based on content analysis and their effectiveness is susceptible to interference from various image or video post-processing operations.Recent research has highlighted the potential of file containers analysis as a promising forensic approach that offers efficient and interpretable results.However,there is still a lack of review articles on this kind of approach.In order to fill this gap,we present a comprehensive review of file containers-based image and video forensics in this paper.Specifically,we categorize the existing methods into two distinct stages,qualitative analysis and quantitative analysis.In addition,an overall framework is proposed to organize the exiting approaches.Then,the advantages and disadvantages of the schemes used across different forensic tasks are provided.Finally,we outline the trends in this research area,aiming to provide valuable insights and technical guidance for future research.
文摘Artifcial Intelligence (AI) is being applied to improve the efciency of software systems used in various domains, especially in the health and forensic sciences. Explainable AI (XAI) is one of the felds of AI that interprets and explains the methods used in AI. One of the techniques used in XAI to provide such interpretations is by computing the rel-evanceof the input features to the output of an AI model. File fragment classifcation is one of the vital issues of fle carving in Cyber Forensics (CF) and becomes challenging when the flesystem metadata is missing. Other major challenges it faces are: proliferation of fle formats, fle embeddings, automation, We leverage and utilize interpretations provided by XAI to optimize the classifcation of fle fragments and propose a novel sifting approach, named SIFT (Sifting File Types). SIFT employs TF-IDF to assign weight to a byte (feature), which is used to select features from a fle fragment. Threshold-based LIME and SHAP (the two XAI techniques) feature relevance values are computed for the selected features to optimize fle fragment classifcation. To improve multinomial classifcation, a Multilayer Per-ceptronmodel is developed and optimized with fve hidden layers, each layer with i × n neurons, where i = the layer number and n = the total number of classes in the dataset. When tested with 47,482 samples of 20 fle types (classes), SIFT achieves a detection rate of 82.1% and outperforms the other state-of-the-art techniques by at least 10%. To the best of our knowledge, this is the frst efort of applying XAI in CF for optimizing fle fragment classifcation.
文摘Electric Vehicle Charging Systems(EVCS)are increasingly vulnerable to cybersecurity threats as they integrate deeply into smart grids and Internet ofThings(IoT)environments,raising significant security challenges.Most existing research primarily emphasizes network-level anomaly detection,leaving critical vulnerabilities at the host level underexplored.This study introduces a novel forensic analysis framework leveraging host-level data,including system logs,kernel events,and Hardware Performance Counters(HPC),to detect and analyze sophisticated cyberattacks such as cryptojacking,Denial-of-Service(DoS),and reconnaissance activities targeting EVCS.Using comprehensive forensic analysis and machine learning models,the proposed framework significantly outperforms existing methods,achieving an accuracy of 98.81%.The findings offer insights into distinct behavioral signatures associated with specific cyber threats,enabling improved cybersecurity strategies and actionable recommendations for robust EVCS infrastructure protection.
基金supported by the Key Research and Development Program of Shaanxi(No.S2024-YF-YB-SF-1359).
文摘Objective Current autosomal short tandem repeat(STR)assays can analyze the zygotic composition by comparing the allelic genes at each locus of complete hydatidiform moles(CHM),with a maternal genotype serving as an essential reference for comparative analysis.However,their application in pathology represents a challenge because of deficiency or contamination of maternal-origin tissues.This study aimed to develop a novel STR genotyping method for identifying CHM genotypes without a maternal component.Methods Samples with the pathologic description of molar pregnancy were collected.Routine hematoxylin–eosin(HE)staining and p57 immunohistochemistry staining were conducted in accordance with standard guidelines.A novel 26-plex system was explored to classify CHM and diploid pregnancies.The system combined 22 STRs on chromosomes 21/18/13/X,3 sex loci,and 1 quality control marker(TAF9L),enabling molecular diagnosis in the absence of maternal tissue.At last,traditional DNA typing based on villi and decidua(maternal component)of each case was used for result consistency analysis.Results CHM and nonmolar abortus could not be distinguished by the basic HE staining with no fetal evidence or other prominent features.DNA typing was successfully processed for all cases according to the novel 26-plex and traditional system.CHM(46XX)diagnosis required single A-STR/X-STR peaks and absent Y-chromosome markers,excluding chromosomal abnormalities via TAF9L analysis.When the villous tissue analysis revealed single peaks at X-STR/SRY loci,a 1:1 amelogenin ratio,and a 2:1 TAF9L peak ratio,these results overlapped with those of 46XY hydropic abortus or CHM.Notably,p57 immunohistochemical staining resolved the ambiguity.Consistency with traditional DNA genotyping confirmed system accuracy.This multiplex assay enhanced reliability in mole diagnosis,supporting clinical differentiation and genetic counseling.Conclusion This study presents a rapid and cost-effective assay for the genotypic identification of CHM without the need for a maternal component.The method combined the characteristics of STR loci distributed across different chromosomes and developed the clinic application of forensic biomarkers.
基金supported by Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2021R1I1A3049788).
文摘In today’s digital era,the rapid evolution of image editing technologies has brought about a significant simplification of image manipulation.Unfortunately,this progress has also given rise to the misuse of manipulated images across various domains.One of the pressing challenges stemming from this advancement is the increasing difficulty in discerning between unaltered and manipulated images.This paper offers a comprehensive survey of existing methodologies for detecting image tampering,shedding light on the diverse approaches employed in the field of contemporary image forensics.The methods used to identify image forgery can be broadly classified into two primary categories:classical machine learning techniques,heavily reliant on manually crafted features,and deep learning methods.Additionally,this paper explores recent developments in image forensics,placing particular emphasis on the detection of counterfeit colorization.Image colorization involves predicting colors for grayscale images,thereby enhancing their visual appeal.The advancements in colorization techniques have reached a level where distinguishing between authentic and forged images with the naked eye has become an exceptionally challenging task.This paper serves as an in-depth exploration of the intricacies of image forensics in the modern age,with a specific focus on the detection of colorization forgery,presenting a comprehensive overview of methodologies in this critical field.
文摘Advances in technological developments in Geographic Information Systems(GIS)has enabled the application of GIS in landscape mapping,environmental management,natural hazard risk and disaster management.As geographical information becomes more widely available through satellite and aerial imagery,the cost of software decreases and GIS expertise expands,it is most likely that the use of GIS will increase.The methodology has practical applications for police,crime scene investigators and forensic geoscientists.The aim is to develop GIS use in forensic search beyond mapping to offer a set of decision support tools that utilise the spatial analytical capabilities of GIS.This enables better management and understanding of the complicated and interrelated nature of a ground search.
基金National Natural Science Foundation of China(No.61602109)Distinguished Young Professor Program of Donghua University,China(No.LZB2019003)+1 种基金Shanghai Science and Technology Innovation Action Plan,China(No.19511101802)Fundamental Research Funds for the Central Universities。
文摘In this age when most organizations make use of cloud computing,it is important to not only protect cloud computing resources from cyber⁃attacks but also investigate these attacks.During forensic investigations in a cloud environment,the investigators fall on service providers for pieces of evidence like log files.The challenge,however,is the integrity of these logs provided by the service providers.To this end,we propose a blockchain⁃based log verification system called BlogVerifier that uses a decentralized approach to solve forensics issues in the cloud.BlogVerifier extracts logs produced in cloud environments,hashes these logs and stores the hashed values as transactional values on the blockchain.The transactions are then merged into blocks and shared on the blockchain.The proposed system also ensures the continuation of an investigation even when the primary source of a log is compromised by using encryption and smart contracts.The proposed system also makes it possible for any stakeholder involved in the forensic process to verify the authenticity of log files.The performance results show that BlogVerifier can be integrated into the cloud environment without any significant impact on system resources and increase in computational cost.
文摘The proliferation of cloud computing and internet of things has led to the connectivity of states and nations(developed and developing countries)worldwide in which global network provide platform for the connection.Digital forensics is a field of computer security that uses software applications and standard guidelines which support the extraction of evidences from any computer appliances which is perfectly enough for the court of law to use and make a judgment based on the comprehensiveness,authenticity and objectivity of the information obtained.Cybersecurity is of major concerned to the internet users worldwide due to the recent form of attacks,threat,viruses,intrusion among others going on every day among internet of things.However,it is noted that cybersecurity is based on confidentiality,integrity and validity of data.The aim of this work is make a systematic review on the application of machine learning algorithms to cybersecurity and cyber forensics and pave away for further research directions on the application of deep learning,computational intelligence,soft computing to cybersecurity and cyber forensics.
基金supported by the MSIT(Ministry of Science and ICT),Korea,under the ITRC(Information Technology Research Center)support program(IITP-2024-RS-2024-00437494)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation).
文摘Digital forensics aims to uncover evidence of cybercrimes within compromised systems.These cybercrimes are often perpetrated through the deployment of malware,which inevitably leaves discernible traces within the compromised systems.Forensic analysts are tasked with extracting and subsequently analyzing data,termed as artifacts,from these systems to gather evidence.Therefore,forensic analysts must sift through extensive datasets to isolate pertinent evidence.However,manually identifying suspicious traces among numerous artifacts is time-consuming and labor-intensive.Previous studies addressed such inefficiencies by integrating artificial intelligence(AI)technologies into digital forensics.Despite the efforts in previous studies,artifacts were analyzed without considering the nature of the data within them and failed to prove their efficiency through specific evaluations.In this study,we propose a system to prioritize suspicious artifacts from compromised systems infected with malware to facilitate efficient digital forensics.Our system introduces a double-checking method that recognizes the nature of data within target artifacts and employs algorithms ideal for anomaly detection.The key ideas of this method are:(1)prioritize suspicious artifacts and filter remaining artifacts using autoencoder and(2)further prioritize suspicious artifacts and filter remaining artifacts using logarithmic entropy.Our evaluation demonstrates that our system can identify malicious artifacts with high accuracy and that its double-checking method is more efficient than alternative approaches.Our system can significantly reduce the time required for forensic analysis and serve as a reference for future studies.
文摘Detecting hate speech automatically in social media forensics has emerged as a highly challenging task due tothe complex nature of language used in such platforms. Currently, several methods exist for classifying hatespeech, but they still suffer from ambiguity when differentiating between hateful and offensive content and theyalso lack accuracy. The work suggested in this paper uses a combination of the Whale Optimization Algorithm(WOA) and Particle Swarm Optimization (PSO) to adjust the weights of two Multi-Layer Perceptron (MLPs)for neutrosophic sets classification. During the training process of the MLP, the WOA is employed to exploreand determine the optimal set of weights. The PSO algorithm adjusts the weights to optimize the performanceof the MLP as fine-tuning. Additionally, in this approach, two separate MLP models are employed. One MLPis dedicated to predicting degrees of truth membership, while the other MLP focuses on predicting degrees offalse membership. The difference between these memberships quantifies uncertainty, indicating the degree ofindeterminacy in predictions. The experimental results indicate the superior performance of our model comparedto previous work when evaluated on the Davidson dataset.
文摘The AGCU X Plus STR system is a newly developed multiplex PCR kit that detects 32 X-chromosomal STR loci simultaneously.These are DXS6807,DXS9895,linkage group 1(DXS10148,DXS10135,DXS8378),DXS9902,DXS6795,DXS6810,DXS10159,DXS10162,DXS10164,DXS7132,linkage group 2(DXS10079,DXS10074,DXS10075),DXS981,DXS6800,DXS6803,DXS6809,DXS6789,DXS7424,DXS101,DXS7133,GATA172D05,GATA165B12,linkage group 3(DXS10103,HPRTB,DXS10101),GATA31E08 and linkage group 4(DXS8377,DXS10134,DXS7423).A major advantage of this kit is that it takes into account linkage between loci,in addition to detecting more X-STR loci.In order to evaluate the forensic application of 32 X-STR fl uorescence amplifi cation system,PCR settings,sensitivity,species specifi city,stability,DNA mixtures,concordance,stutter,sizing precision,and population genetics investigation were evaluated according to the Scientific Working Group on DNA Analysis Methods(SWGDAM)developmental validation guidelines.The study showed that the genotyping results of each locus were signifi cantly accurate when the DNA template was at least 62.5 pg.Complete profi les were obtained for the 1∶1 and 1∶3 combinations.A total of 209 unrelated individuals from Southern Chinese Han community,consisting of 84 females and 125 males,were selected for population studies,and 285 allele profi les were detected from 32 X-STR loci.The polymorphism information content(PIC)ranged from 0.2721 in DXS6800,to 0.9105 in DXS10135,with an average of 0.6798.DXS10135(PIC=0.9105)was the most polymorphic locus,with discrimination power(DP)of 0.9164 and 0.9871 for the male and female.The cumulative PD_(F),PD_(M),MEC_(trio) and MEC_(duo) valu es were all greater than 0.999999999.There were 78 different DXS10103-HPRTB-DXS10101 haplotypes among the 125 males,and the haplotype diversity was 0.9810.There was no signifi cant difference in the cumulative PD_(F),PD_(M),MEC_(trio) and MEC_(duo) values whether considering linkage or not.In summary,the new X-STR multiplex typing system is effective and reliable,which can be useful in human genetic analysis and kinship testing as a potent complement to autosomal STR typing.
文摘Signature verification involves vague situations in which a signature could resemble many reference samples ormight differ because of handwriting variances. By presenting the features and similarity score of signatures from thematching algorithm as fuzzy sets and capturing the degrees of membership, non-membership, and indeterminacy,a neutrosophic engine can significantly contribute to signature verification by addressing the inherent uncertaintiesand ambiguities present in signatures. But type-1 neutrosophic logic gives these membership functions fixed values,which could not adequately capture the various degrees of uncertainty in the characteristics of signatures. Type-1neutrosophic representation is also unable to adjust to various degrees of uncertainty. The proposed work exploresthe type-2 neutrosophic logic to enable additional flexibility and granularity in handling ambiguity, indeterminacy,and uncertainty, hence improving the accuracy of signature verification systems. Because type-2 neutrosophiclogic allows the assessment of many sources of ambiguity and conflicting information, decision-making is moreflexible. These experimental results show the possible benefits of using a type-2 neutrosophic engine for signatureverification by demonstrating its superior handling of uncertainty and variability over type-1, which eventuallyresults in more accurate False Rejection Rate (FRR) and False Acceptance Rate (FAR) verification results. In acomparison analysis using a benchmark dataset of handwritten signatures, the type-2 neutrosophic similaritymeasure yields a better accuracy rate of 98% than the type-1 95%.
基金President’s Fund Natural Science Project Plan of Tarim University,“Research on Blockchain-based Electronic Evidence Fixation Method”(Project No.:TDZKSS202439)。
文摘The development of high technology,for public life to provide a justification at the same time,also encouraged the spirit of cybercrime,to become more and more rampant.In network crime,electronic data is usually used as the main evidence to determine the facts of the crime and plays an important role in the smooth trial of the case.But because electronic data on dependent,concealment,easy destructive strong science and technology,the forensics work is now in trouble.The mature use of blockchain technology can avoid existing problems to a certain extent,which is helpful to the smooth progress of electronic forensics.This paper on electronic evidence how to more effectively,combined with research blockchain technology,improve the efficiency of electronic evidence collection work.
基金QuỹĐổi mới sáng tạo Vingroup,Grant/Award Number:VINIF.2020.ThS.BK.10。
文摘In recent years,visual facial forgery has reached a level of sophistication that humans cannot identify fraud,which poses a significant threat to information security.A wide range of malicious applications have emerged,such as deepfake,fake news,defamation or blackmailing of celebrities,impersonation of politicians in political warfare,and the spreading of rumours to attract views.As a result,a rich body of visual forensic techniques has been proposed in an attempt to stop this dangerous trend.However,there is no comprehensive,fair,and unified performance evaluation to enlighten the community on best performing methods.The authors present a systematic benchmark beyond traditional surveys that provides in-depth insights into facial forgery and facial forensics,grounding on robustness tests such as contrast,brightness,noise,resolution,missing information,and compression.The authors also provide a practical guideline of the benchmarking results,to determine the characteristics of the methods that serve as a comparative reference in this never-ending war between measures and countermeasures.The authors’source code is open to the public.
基金the National Key R&D Program of China(No.2017YFA60700602)。
文摘Vehicle data is one of the important sources of traffic accident digital forensics.We propose a novel method using long short-term memory-deep belief network by binary encoding(LSTM-BiDBN)controller area network identifier(CAN ID)to extract the event sequence of CAN IDs and the semantic of CAN IDs themselves.Instead of detecting attacks only aimed at a specific CAN ID,the proposed method fully considers the potential interaction between electronic control units.By this means,we can detect whether the vehicle has been invaded by the outside,to online determine the responsible party of the accident.We use our LSTM-BiDBN to distinguish attack-free and abnormal situations on CAN-intrusion-dataset.Experimental results show that our proposed method is more effective in identifying anomalies caused by denial of service attack,fuzzy attack and impersonation attack with an accuracy value of 97.02%,a false-positive rate of 6.09%,and a false-negative rate of 1.94%compared with traditional methods.
基金supported by National Natural Science Foundation of China(No.61502274).
文摘In the field of image forensics,image tampering detection is a critical and challenging task.Traditional methods based on manually designed feature extraction typically focus on a specific type of tampering operation,which limits their effectiveness in complex scenarios involving multiple forms of tampering.Although deep learningbasedmethods offer the advantage of automatic feature learning,current approaches still require further improvements in terms of detection accuracy and computational efficiency.To address these challenges,this study applies the UNet 3+model to image tampering detection and proposes a hybrid framework,referred to as DDT-Net(Deep Detail Tracking Network),which integrates deep learning with traditional detection techniques.In contrast to traditional additive methods,this approach innovatively applies amultiplicative fusion technique during downsampling,effectively combining the deep learning feature maps at each layer with those generated by the Bayar noise stream.This design enables noise residual features to guide the learning of semantic features more precisely and efficiently,thus facilitating comprehensive feature-level interaction.Furthermore,by leveraging the complementary strengths of deep networks in capturing large-scale semantic manipulations and traditional algorithms’proficiency in detecting fine-grained local traces,the method significantly enhances the accuracy and robustness of tampered region detection.Compared with other approaches,the proposed method achieves an F1 score improvement exceeding 30% on the DEFACTO and DIS25k datasets.In addition,it has been extensively validated on other datasets,including CASIA and DIS25k.Experimental results demonstrate that this method achieves outstanding performance across various types of image tampering detection tasks.
文摘Recent advances in artificial intelligence and the availability of large-scale benchmarks have made deepfake video generation and manipulation easier.Therefore,developing reliable and robust deepfake video detection mechanisms is paramount.This research introduces a novel real-time deepfake video detection framework by analyzing gaze and blink patterns,addressing the spatial-temporal challenges unique to gaze and blink anomalies using the TimeSformer and hybrid Transformer-CNN models.The TimeSformer architecture leverages spatial-temporal attention mechanisms to capture fine-grained blinking intervals and gaze direction anomalies.Compared to state-of-the-art traditional convolutional models like MesoNet and EfficientNet,which primarily focus on global facial features,our approach emphasizes localized eye-region analysis,significantly enhancing detection accuracy.We evaluate our framework on four standard datasets:FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb.The proposed framework results reveal higher accuracy,with the TimeSformer model achieving accuracies of 97.5%,96.3%,95.8%,and 97.1%,and with the hybrid Transformer-CNN model demonstrating accuracies of 92.8%,91.5%,90.9%,and 93.2%,on FaceForensics,CelebDF-V2,DFDC,and FakeAVCeleb datasets,respectively,showing robustness in distinguishing manipulated from authentic videos.Our research provides a robust state-of-the-art framework for real-time deepfake video detection.This novel study significantly contributes to video forensics,presenting scalable and accurate real-world application solutions.
