{"title":"RIS-Assisted Integrated Communication and Secret Key Generation in Quasi-Static Environments","authors":"Zheyuan Deng, Xiaoyan Hu, Keming Ma, Liang Jin, Boming Li, Jinghua Qu","doi":"10.1109/tifs.2026.3654398","DOIUrl":"https://doi.org/10.1109/tifs.2026.3654398","url":null,"abstract":"","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"267 1","pages":""},"PeriodicalIF":6.8,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145972226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-14DOI: 10.1109/tifs.2026.3654388
Sk Tanzir Mehedi, Chadni Islam, Gowri Ramachandran, Raja Jurdak
{"title":"DySec: A Machine Learning-based Dynamic Analysis for Detecting Malicious Packages in PyPI Ecosystem","authors":"Sk Tanzir Mehedi, Chadni Islam, Gowri Ramachandran, Raja Jurdak","doi":"10.1109/tifs.2026.3654388","DOIUrl":"https://doi.org/10.1109/tifs.2026.3654388","url":null,"abstract":"","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"5 1","pages":""},"PeriodicalIF":6.8,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145972220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vertical Split Federated Learning (VSFL) allows participants to collaboratively train a better model with different features vertically partitioned in the same sample space, where the model is divided into bottom model and top model by the cut layer, trained by passive and active participants respectively. However, in the process, the labels owned by the active participant will still be inferred or stolen by curious or malicious passive participants. In this paper, we propose Casper, a causality-inspired defense mechanism with a confounder against label inference attacks in VSFL. Casper first analyzes the feasibility of optimizing the training process in VSFL at the intervention level from a causal perspective. It then introduces a confounder consisting of cut layer output reconstruction and label obfuscation to disrupt the direct causality between cut layer outputs and labels. Additionally, we integrate selective discrepancy training to further ensure model utility by strategically balancing training between active and passive participants. Extensive experiments conducted on four datasets across different tasks demonstrate that Casper effectively preserves label privacy while maintaining model performance, significantly outperforming current advanced defending methods in VSFL.
{"title":"Casper: A Causality-Inspired Defense With Confounder Against Label Inference Attacks in Vertical Split Federated Learning","authors":"Meng Shen;Jin Meng;Bohan Peng;Xiangyun Tang;Wei Wang;Dusit Niyato;Liehuang Zhu","doi":"10.1109/TIFS.2026.3652013","DOIUrl":"https://doi.org/10.1109/TIFS.2026.3652013","url":null,"abstract":"Vertical Split Federated Learning (VSFL) allows participants to collaboratively train a better model with different features vertically partitioned in the same sample space, where the model is divided into bottom model and top model by the cut layer, trained by passive and active participants respectively. However, in the process, the labels owned by the active participant will still be inferred or stolen by curious or malicious passive participants. In this paper, we propose Casper, a causality-inspired defense mechanism with a confounder against label inference attacks in VSFL. Casper first analyzes the feasibility of optimizing the training process in VSFL at the intervention level from a causal perspective. It then introduces a confounder consisting of cut layer output reconstruction and label obfuscation to disrupt the direct causality between cut layer outputs and labels. Additionally, we integrate selective discrepancy training to further ensure model utility by strategically balancing training between active and passive participants. Extensive experiments conducted on four datasets across different tasks demonstrate that Casper effectively preserves label privacy while maintaining model performance, significantly outperforming current advanced defending methods in VSFL.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"1050-1064"},"PeriodicalIF":8.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-12DOI: 10.1109/TIFS.2026.3651997
Di Li;Chun Li;Yufeng Tang;Yupeng Zhang;Zheng Gong
Inconsistencies in manufacturing features, sampling settings, and cryptographic implementations amongst the profiling and target devices can lead to the failure of profiling side-channel analysis (SCA). Various techniques, such as preprocessing, multi-device training, and transfer learning, have been proposed to mitigate this portability problem in profiling SCA. However, many techniques of block ciphers, such as tweaks, key-dependent components, and customized elements, might have uncertain effects from the perspective of cryptographic implementations, requiring further insightful analysis on their impact on portability. This paper investigates the portability of profiling SCA from a case study using adjustable implementations of block ciphers. First, we theoretically analyze the variation in leakage distribution under adjustable implementations. To support our theoretical results, a dataset of deep-learning SCA is built from AES, Pilsung, and Skinny. Specifically, we reveal how to reverse the parameterized components and recover the key from these adjustable implementations. According to our experiment on an 8-bit AVR microcontroller, the computational complexities of the attacks based on our model are less than $9times 2^{16}$ within 4500 traces. Moreover, the effectiveness of our proposed method is demonstrated under the combinatorial effect with adjustable implementations and device characteristics. Our case study provides insights into the results of adjustable implementations of block ciphers, which strengthens both the theoretical and practical understanding of the portability of profiling SCA.
{"title":"Portability of Profiling Side-Channel Analysis: A Case Study Using Adjustable Implementations of Block Ciphers","authors":"Di Li;Chun Li;Yufeng Tang;Yupeng Zhang;Zheng Gong","doi":"10.1109/TIFS.2026.3651997","DOIUrl":"https://doi.org/10.1109/TIFS.2026.3651997","url":null,"abstract":"Inconsistencies in manufacturing features, sampling settings, and cryptographic implementations amongst the profiling and target devices can lead to the failure of profiling side-channel analysis (SCA). Various techniques, such as preprocessing, multi-device training, and transfer learning, have been proposed to mitigate this portability problem in profiling SCA. However, many techniques of block ciphers, such as tweaks, key-dependent components, and customized elements, might have uncertain effects from the perspective of cryptographic implementations, requiring further insightful analysis on their impact on portability. This paper investigates the portability of profiling SCA from a case study using adjustable implementations of block ciphers. First, we theoretically analyze the variation in leakage distribution under adjustable implementations. To support our theoretical results, a dataset of deep-learning SCA is built from AES, Pilsung, and Skinny. Specifically, we reveal how to reverse the parameterized components and recover the key from these adjustable implementations. According to our experiment on an 8-bit AVR microcontroller, the computational complexities of the attacks based on our model are less than <inline-formula> <tex-math>$9times 2^{16}$ </tex-math></inline-formula> within 4500 traces. Moreover, the effectiveness of our proposed method is demonstrated under the combinatorial effect with adjustable implementations and device characteristics. Our case study provides insights into the results of adjustable implementations of block ciphers, which strengthens both the theoretical and practical understanding of the portability of profiling SCA.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"1021-1035"},"PeriodicalIF":8.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145982335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-12DOI: 10.1109/TIFS.2026.3652010
Chen Zhu;Yang Lu;Nian Xia;Jiguo Li;Yinxia Sun
Data Integrity Auditing (DIA) enables users to remotely verify whether their data saved in third-party clouds has been maliciously tampered with or compromised. As an extension of DIA in certificateless cryptography, certificateless DIA (CL-DIA) integrates the merits of conventional public-key cryptography (no key escrow) and identity-based cryptography (no certificates). However, CL-DIA schemes depend on a reliable third-party auditor (TPA) to perform integrity audits, inevitably suffering from performance bottleneck and single-point failure problems. Moreover, almost all current CL-DIA schemes were designed with computationally expensive bilinear pairings. Cryptanalysis demonstrates that the existing unique pairing-free CL-DIA scheme fails to achieve the unforgeable security of auditing proofs. In this work, we put forward a lightweight blockchain-assisted CL-DIA scheme. The scheme achieves DIA through the blockchain instead of a single TPA, thereby overcoming the problems caused by the TPA-based centralized auditing model. Then, by avoiding time-consuming pairing operations and employing edge servers in generating verifiable tags for the uploaded data of users, its performance surpasses previous pairing-based CL-DIA schemes, particularly in terms of computation efficiency. Furthermore, we provide formal proofs in the random oracle model demonstrating that our scheme achieves unforgeability of verifiable tags and auditing proofs, ensures data privacy secrity, and is resistant to collusion attacks between the EN and the CSP. Finally, experimental results show that when auditing 25 file blocks, our scheme only costs 0.29s, which reduces the total time cost of integrity auditing phase by 48.2%-85.5% compared to current pairing-based CL-DIA schemes.
{"title":"A Lightweight Blockchain-Assisted Certificateless Cloud Data Integrity Auditing Scheme Without Third-Party Auditor","authors":"Chen Zhu;Yang Lu;Nian Xia;Jiguo Li;Yinxia Sun","doi":"10.1109/TIFS.2026.3652010","DOIUrl":"https://doi.org/10.1109/TIFS.2026.3652010","url":null,"abstract":"Data Integrity Auditing (DIA) enables users to remotely verify whether their data saved in third-party clouds has been maliciously tampered with or compromised. As an extension of DIA in certificateless cryptography, certificateless DIA (CL-DIA) integrates the merits of conventional public-key cryptography (no key escrow) and identity-based cryptography (no certificates). However, CL-DIA schemes depend on a reliable third-party auditor (TPA) to perform integrity audits, inevitably suffering from performance bottleneck and single-point failure problems. Moreover, almost all current CL-DIA schemes were designed with computationally expensive bilinear pairings. Cryptanalysis demonstrates that the existing unique pairing-free CL-DIA scheme fails to achieve the unforgeable security of auditing proofs. In this work, we put forward a lightweight blockchain-assisted CL-DIA scheme. The scheme achieves DIA through the blockchain instead of a single TPA, thereby overcoming the problems caused by the TPA-based centralized auditing model. Then, by avoiding time-consuming pairing operations and employing edge servers in generating verifiable tags for the uploaded data of users, its performance surpasses previous pairing-based CL-DIA schemes, particularly in terms of computation efficiency. Furthermore, we provide formal proofs in the random oracle model demonstrating that our scheme achieves unforgeability of verifiable tags and auditing proofs, ensures data privacy secrity, and is resistant to collusion attacks between the EN and the CSP. Finally, experimental results show that when auditing 25 file blocks, our scheme only costs 0.29s, which reduces the total time cost of integrity auditing phase by 48.2%-85.5% compared to current pairing-based CL-DIA schemes.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"976-991"},"PeriodicalIF":8.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-12DOI: 10.1109/TIFS.2026.3653569
Fan Zhang;Chen Shao;Kangning Du;Yanan Guo;Peiran Song;Lin Cao;Xin Yuan
In recent years, contrastive learning has made significant progress in DeepFake detection. However, existing methods emphasize class granularity, and it is difficult to distinguish between the real instance and its forgery counterparts effectively. Furthermore, the diversity of forgery cues produced by different manipulation methods cannot be effectively clustered by class granularity alone. Thus, the model’s generalization capability is limited. To tackle the above problems, a Dual-Granularity Contrastive Learning (DGCL) for DeepFake detection is proposed in this paper. Specifically, Class Granularity Contrastive Learning (CGCL) and Instance Granularity Contrastive Learning (IGCL) are designed. Firstly, for semantic aggregation at the class level, CGCL incorporates the class prototype, which encourages anchor approaches to the prototype of the positive class, thereby pulling the intra-class features closer. Secondly, for distinguishing between real and fake instances, Real Instance Granularity Contrastive Learning (RIGCL) and Fake Instance Granularity Contrastive Learning (FIGCL) are proposed based on the instance characteristics. RIGCL endeavors to distinguish fake instances from original real instances by expanding the differentiation in the feature space. Meanwhile, FIGCL extracts consistent forgery features from various manipulation methods using cosine similarity constraints. Finally, the superiority and generalizability of DGCL are validated by the experimental results on CELEBDF, DFD, and DFDC datasets.
{"title":"Dual-Granularity Contrastive Learning for DeepFake Detection","authors":"Fan Zhang;Chen Shao;Kangning Du;Yanan Guo;Peiran Song;Lin Cao;Xin Yuan","doi":"10.1109/TIFS.2026.3653569","DOIUrl":"https://doi.org/10.1109/TIFS.2026.3653569","url":null,"abstract":"In recent years, contrastive learning has made significant progress in DeepFake detection. However, existing methods emphasize class granularity, and it is difficult to distinguish between the real instance and its forgery counterparts effectively. Furthermore, the diversity of forgery cues produced by different manipulation methods cannot be effectively clustered by class granularity alone. Thus, the model’s generalization capability is limited. To tackle the above problems, a Dual-Granularity Contrastive Learning (DGCL) for DeepFake detection is proposed in this paper. Specifically, Class Granularity Contrastive Learning (CGCL) and Instance Granularity Contrastive Learning (IGCL) are designed. Firstly, for semantic aggregation at the class level, CGCL incorporates the class prototype, which encourages anchor approaches to the prototype of the positive class, thereby pulling the intra-class features closer. Secondly, for distinguishing between real and fake instances, Real Instance Granularity Contrastive Learning (RIGCL) and Fake Instance Granularity Contrastive Learning (FIGCL) are proposed based on the instance characteristics. RIGCL endeavors to distinguish fake instances from original real instances by expanding the differentiation in the feature space. Meanwhile, FIGCL extracts consistent forgery features from various manipulation methods using cosine similarity constraints. Finally, the superiority and generalizability of DGCL are validated by the experimental results on CELEBDF, DFD, and DFDC datasets.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"1552-1565"},"PeriodicalIF":8.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-12DOI: 10.1109/TIFS.2026.3653580
Xunqiang Lan;Xiao Tang;Ruonan Zhang;Bin Li;Qinghe Du;Dusit Niyato;Zhu Han
Blockchain plays a crucial role in ensuring the security and integrity of decentralized systems, with the proof-of-work (PoW) mechanism being fundamental for achieving distributed consensus. As PoW blockchains see broader adoption, an increasingly diverse set of miners with varying computing capabilities participate in the network. In this paper, we consider the PoW blockchain mining, where the miners are associated with resource uncertainties. To characterize the uncertainty computing resources at different mining participants, we establish an ambiguous set representing uncertainty of resource distributions. Then, the networked mining is formulated as a non-cooperative game, where distributionally robust performance is calculated for each individual miner to tackle the resource uncertainties. We prove the existence of the equilibrium of the distributionally robust mining game. To derive the equilibrium, we propose the conditional value-at-risk (CVaR)-based reinterpretation of the best response of each miner. We then solve the individual strategy with alternating optimization, which facilitates the iteration among miners towards the game equilibrium. Furthermore, we consider the case that the ambiguity of resource distribution reduces to Gaussian distribution and the case that another uncertainties vanish, and then characterize the properties of the equilibrium therein along with a distributed algorithm to achieve the equilibrium. Simulation results show that the proposed approaches effectively converge to the equilibrium, and effectively tackle the uncertainties in blockchain mining to achieve a robust performance guarantee.
{"title":"Distributionally Robust Game for Proof-of-Work Blockchain Mining Under Resource Uncertainties","authors":"Xunqiang Lan;Xiao Tang;Ruonan Zhang;Bin Li;Qinghe Du;Dusit Niyato;Zhu Han","doi":"10.1109/TIFS.2026.3653580","DOIUrl":"https://doi.org/10.1109/TIFS.2026.3653580","url":null,"abstract":"Blockchain plays a crucial role in ensuring the security and integrity of decentralized systems, with the proof-of-work (PoW) mechanism being fundamental for achieving distributed consensus. As PoW blockchains see broader adoption, an increasingly diverse set of miners with varying computing capabilities participate in the network. In this paper, we consider the PoW blockchain mining, where the miners are associated with resource uncertainties. To characterize the uncertainty computing resources at different mining participants, we establish an ambiguous set representing uncertainty of resource distributions. Then, the networked mining is formulated as a non-cooperative game, where distributionally robust performance is calculated for each individual miner to tackle the resource uncertainties. We prove the existence of the equilibrium of the distributionally robust mining game. To derive the equilibrium, we propose the conditional value-at-risk (CVaR)-based reinterpretation of the best response of each miner. We then solve the individual strategy with alternating optimization, which facilitates the iteration among miners towards the game equilibrium. Furthermore, we consider the case that the ambiguity of resource distribution reduces to Gaussian distribution and the case that another uncertainties vanish, and then characterize the properties of the equilibrium therein along with a distributed algorithm to achieve the equilibrium. Simulation results show that the proposed approaches effectively converge to the equilibrium, and effectively tackle the uncertainties in blockchain mining to achieve a robust performance guarantee.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"1036-1049"},"PeriodicalIF":8.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145982203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anonymous submissions inspire people to speak up since hiding their identities can protect them from negative influence by their own words. However, the abuse of anonymity may bring harassment to those public submission callers. Existing works only handle DoS attacks or block harassment senders in an active manner, which behave poorly in the early prevention of uncharacterized harassment. In this paper, we propose MsgFliter, a sender-anonymous messaging system with proactive anti-harassment mechanism. Our core idea is to prevent unanswered senders from sending messages continually while keeping their identities, messages, and sender types secret. To meet the functionality and security requirements of MsgFliter, we propose the Anti-Harassment Anonymous Authentication (AHAA) protocol. We associate messages from the same sender through linkable tags and invalidate linkability when a message is replied to. To achieve session indistinguishability, we further combine the proposed anonymous authentication with zero-knowledge proofs of disjunctive relations. We implement MsgFliter and compare its performance with related solutions. Experimental results show that our solution is practicable.
{"title":"MsgFilter: Proactive Anti-Harassment Sender-Anonymous Messaging System","authors":"Siqin Li;Kun He;Min Shi;Yajing Huang;Ruiying Du;Jing Chen","doi":"10.1109/TIFS.2025.3650417","DOIUrl":"10.1109/TIFS.2025.3650417","url":null,"abstract":"Anonymous submissions inspire people to speak up since hiding their identities can protect them from negative influence by their own words. However, the abuse of anonymity may bring harassment to those public submission callers. Existing works only handle DoS attacks or block harassment senders in an active manner, which behave poorly in the early prevention of uncharacterized harassment. In this paper, we propose MsgFliter, a sender-anonymous messaging system with proactive anti-harassment mechanism. Our core idea is to prevent unanswered senders from sending messages continually while keeping their identities, messages, and sender types secret. To meet the functionality and security requirements of MsgFliter, we propose the Anti-Harassment Anonymous Authentication (AHAA) protocol. We associate messages from the same sender through linkable tags and invalidate linkability when a message is replied to. To achieve session indistinguishability, we further combine the proposed anonymous authentication with zero-knowledge proofs of disjunctive relations. We implement MsgFliter and compare its performance with related solutions. Experimental results show that our solution is practicable.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"741-756"},"PeriodicalIF":8.0,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145893717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-02DOI: 10.1109/TIFS.2025.3650391
Caiping Yan;Zhi Lan;Hong Li;Yuqi Li;Zonglin Meng
Non-contact palm-vein recognition has been widely adopted in security-critical applications owing to its contact-free acquisition paradigm and exceptional discriminative power. However, these systems remain susceptible to a spectrum of presentation attacks (PAs), creating significant security risks that require urgent mitigation. Progress on palm-vein anti-spoofing is currently impeded by three fundamental gaps: 1) the absence of an open-source, end-to-end pipeline for preprocessing liveness-detection data; 2) a lack of publicly available datasets specifically tailored to anti-spoofing evaluation; and 3) the unavailability of benchmark studies employing standardized protocols and reference implementations. In light of these key issues, we make the following four contributions. Firstly, we propose a new open-source preprocessing pipeline that can significantly improve model performance, reducing errors by up to 53.3%. Secondly, we introduce PVASD, a new largest known dataset comprised of 1,187,519 images belonging to 5,515 subjects, which consists of 880,241 live palm vein images along with 307,278 spoof attack images including 16 different attack types–both 2d (printed stuff) and 3d (gloves and prosthesis models) attacks captured under a variety of environments using off-the-shelf commercial-grade sensors at five resolutions. Lastly, comprehensive benchmarks are also created by evaluating three types of representative methods namely classical image classification models, face anti-spoofing methods adapted from face domain and anomaly-detection-based approaches, while our experimental results reveal unique characteristics intrinsic only to palm vein spoof attacks, which will hopefully provide valuable guidance to researchers for further investigation. Furthermore, we also expanded PVASD by adding 20,000 spoof samples generated by artificial intelligence, and evaluated the vulnerability of the existing models to deepfake attacks. We will release our preprocessing pipeline, dataset, and benchmark codes at https://github.com/valhongli/PVASD to advance future reproducible studies and accelerate palm-vein anti-spoofing algorithmic research.
{"title":"A Comprehensive Framework for Palm Vein Anti-Spoofing With Preprocessing Pipeline, Dataset, and Benchmark","authors":"Caiping Yan;Zhi Lan;Hong Li;Yuqi Li;Zonglin Meng","doi":"10.1109/TIFS.2025.3650391","DOIUrl":"10.1109/TIFS.2025.3650391","url":null,"abstract":"Non-contact palm-vein recognition has been widely adopted in security-critical applications owing to its contact-free acquisition paradigm and exceptional discriminative power. However, these systems remain susceptible to a spectrum of presentation attacks (PAs), creating significant security risks that require urgent mitigation. Progress on palm-vein anti-spoofing is currently impeded by three fundamental gaps: 1) the absence of an open-source, end-to-end pipeline for preprocessing liveness-detection data; 2) a lack of publicly available datasets specifically tailored to anti-spoofing evaluation; and 3) the unavailability of benchmark studies employing standardized protocols and reference implementations. In light of these key issues, we make the following four contributions. Firstly, we propose a new open-source preprocessing pipeline that can significantly improve model performance, reducing errors by up to 53.3%. Secondly, we introduce PVASD, a new largest known dataset comprised of 1,187,519 images belonging to 5,515 subjects, which consists of 880,241 live palm vein images along with 307,278 spoof attack images including 16 different attack types–both 2d (printed stuff) and 3d (gloves and prosthesis models) attacks captured under a variety of environments using off-the-shelf commercial-grade sensors at five resolutions. Lastly, comprehensive benchmarks are also created by evaluating three types of representative methods namely classical image classification models, face anti-spoofing methods adapted from face domain and anomaly-detection-based approaches, while our experimental results reveal unique characteristics intrinsic only to palm vein spoof attacks, which will hopefully provide valuable guidance to researchers for further investigation. Furthermore, we also expanded PVASD by adding 20,000 spoof samples generated by artificial intelligence, and evaluated the vulnerability of the existing models to deepfake attacks. We will release our preprocessing pipeline, dataset, and benchmark codes at <uri>https://github.com/valhongli/PVASD</uri> to advance future reproducible studies and accelerate palm-vein anti-spoofing algorithmic research.","PeriodicalId":13492,"journal":{"name":"IEEE Transactions on Information Forensics and Security","volume":"21 ","pages":"945-959"},"PeriodicalIF":8.0,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145893716","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：