The DOI matches its Crossref record; 69 of the 69 references it cites were found in a registry or an index; 2 of the 10 sources its claims cite were checked for support.
A Novel Blockchain-Based Deepfake Detection Method Using Federated and Deep Learning Models
Open accessAbstract
In recent years, the proliferation of deep learning (DL) techniques has given rise to a significant challenge in the form of deepfake videos, posing a grave threat to the authenticity of media content. With the rapid advancement of DL technology, the creation of convincingly realistic deepfake videos has become increasingly prevalent, raising serious concerns about the potential misuse of such content. Deepfakes have the potential to undermine trust in visual media, with implications for fields as diverse as journalism, entertainment, and security. This study presents an innovative solution by harnessing blockchain-based federated learning (FL) to address this issue, focusing on preserving data source anonymity. The approach combines the strengths of SegCaps and convolutional neural network (CNN) methods for improved image feature extraction, followed by capsule network (CN) training to enhance generalization. A novel data normalization technique is introduced to tackle data heterogeneity stemming from diverse global data sources. Moreover, transfer learning (TL) and preprocessing methods are deployed to elevate DL performance. These efforts culminate in collaborative global model training zfacilitated by blockchain and FL while maintaining the utmost confidentiality of data sources. The effectiveness of our methodology is rigorously tested and validated through extensive experiments. These experiments reveal a substantial improvement in accuracy, with an impressive average increase of 6.6% compared to six benchmark models. Furthermore, our approach demonstrates a 5.1% enhancement in the area under the curve (AUC) metric, underscoring its ability to outperform existing detection methods. These results substantiate the effectiveness of our proposed solution in countering the proliferation of deepfake content. In conclusion, our innovative approach represents a promising avenue for advancing deepfake detection. By leveraging existing data resources and the power of FL and blockchain technology, we address a critical need for media authenticity and security. As the threat of deepfake videos continues to grow, our comprehensive solution provides an effective means to protect the integrity and trustworthiness of visual media, with far-reaching implications for both industry and society. This work stands as a significant step toward countering the deepfake menace and preserving the authenticity of visual content in a rapidly evolving digital landscape.
Overview
CoChat Summary
TL;DR
The paper proposes BFLDL, a blockchain-based federated learning framework with deep learning for deepfake detection that aims to preserve data-source anonymity while improving accuracy and AUC. It reports accuracy above 96% across four deepfake datasets and average gains of 6.6% in accuracy and 5.1% in AUC over six benchmark models.
Key findings
- BFLDL combines blockchain-based federated learning with SegCaps, CNN, capsule network training, data normalization, and transfer learning for deepfake detection.
- The method reports accuracy and AUC greater than 96% on all examined datasets; on FF++ with DeepFakeTIMIT, accuracy is above 97%, CelebDF reaches 98.9%, and DFDCpre reaches 98.1%.
- Compared with six benchmark models, the approach reports an average accuracy increase of 6.6% and a 5.1% enhancement in AUC.
- Transfer learning improved low-quality video detection accuracy by varying amounts across datasets, increased high-quality DFDC film detection accuracy by 0.4%, and increased AUC in low-quality CelebDF videos by 0.8.
Methods
- Design: A blockchain-based federated learning system where multiple sources/clients train local deep learning models and exchange gradients or model weights through a permissioned blockchain, using PoW consensus and Laplace-based privacy protection; data normalization includes spatial and signal normalization.
- Data: Experiments used FaceForensics++ (FF++), DeepFakeTIMIT, DFDCpre, and CelebDF. FF++ contains more than 1000 deepfake videos per category and compression rates c0, c23, and c40; DFDCpre contains around 5000 movies; videos were framed, faces extracted with DLIB, and frames normalized/scaled to 299×299×3.
- Analysis: Accuracy (ACC) and area under the curve (AUC) were evaluated at frame/image level. Implementation used TensorFlow, Python 3.6, scikit-learn, PySyft, an Intel Core i9 processor, NVIDIA Quadro RTX 6000, and 64 GB RAM; the Adam optimizer was used.
Limitations
- The paper states future work includes real-time deepfake detection using online intrusions that examine video and audio-visual data in 1-s intervals, and combining the method with other techniques such as matrix algebra, Kalman filtering, multitask learning-enabled graph convolution, active subspace random optimization, and deep neural network–based logical and activity learning.
- The text makes evident that running cost rises with the number of clients or transactions because of increased communication overhead, and the total cost of the BFLDL distributed model increased when sources or clients were enhanced.
- The text also makes evident that accuracy does not change smoothly because selection from different resources is not the same, and that blockchain data storage is monetarily and computationally costly due to limited storage space, so original input videos are kept by the sources while blockchain aids in recovering the trained model.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 11 Not valid · 5 Not checked · 52 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“FL allows ML model training while preserving data privacy [35,36,37]”
What the sources say
- Not checked — no full text of the sourcecites [35] Fu Y, et al. An incentive mechanism of incorporating supervision game for federated learning in autonomous driving. IEEE Trans Intell Transp Syst. 2023. source
- Supportedcites [36] Han S, et al. Practical and robust federated learning with highly scalable regression training. IEEE Trans Neural Netw Learn Syst. 2023. source
The Privacy-Preserving Federated Learning section says participants can collaboratively learn a shared model without exposing their local raw data. The paper also describes federated learning as a distributed ML approach in which data owners retain their raw data. (LLM, conf=0.90)
- Not checked — no full text of the sourcecites [37] Wu Z, et al. hPSD: a hybrid PU-learning-based spammer detection model for product reviews. IEEE Trans Cybern. 2018;50(4):1595–606. source
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The detection accuracy of low-quality videos has increased by varying amounts in different datasets after employing the TL technique, as shown in Table 3”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the provided method outperforms the existing cutting-edge deepfake investigative techniques.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The integration of both texture-based analysis and the SegCaps-CNN-based method for feature extraction in the YOLO-based face recognition system stems from their complementary strengths.”
- Also stated in: § Related Work · boosted
“Empirical evidence supports this choice, demonstrating superior performance compared to using either method alone.”
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Related Work · boosted
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The quantity of providers affects the performance of the collaborative model.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
12 more claims
Create a free account to see all 69 claims and every source verdict.
- Probably not a substantial claim. 44
- Borderline: one of the tests for a substantial claim is unclear. 1
- It may state two separate claims as one. 19
- The quote only partly states this claim. 3
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Related Work
“In contrast to Heidari et al.’s work, we focus on lung cancer detection, utilizing the FBCLC-Rad technique, with an emphasis on image preprocessing, spectral analysis–based feature extraction, and CapsNets for classification.”
§ Related Work
“We employ CNN with different architecture and SegCaps to enhance deepfake detection, while Heidari et al. utilized CapsNets for lung cancer diagnosis.”
Citations
Cites
FaceForensics++: Learning to Detect Manipulated Facial Images
Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Niessner, M. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00009 · Cited by 3,314
Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics
Li, Y., Yang, X., Sun, P., Qi, H., & Lyu, S. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00327 · Cited by 1,846 · cited 1×
“FaceForensics (FF + +) [59], DeepFakeTIMIT [60], DeepFakes Detection Challenge Preview (DFDCpre) [61], and CelebDF [62] were the datasets used in the research.”
Multiscale Feature Extraction and Fusion of Image and Text in VQA
Lu, S., Ding, Y., Liu, M., Yin, Z., Yin, L., & Zheng, W. (2023) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-023-00233-6 · Cited by 422
The Deepfake Detection Challenge (DFDC) Preview Dataset
Dolhansky, B., Howes, R., Pflaum, B., Baram, N., & Ferrer, C. C. (2019) · arXiv · doi:10.48550/arxiv.1910.08854 · Cited by 413
Improving Physical Layer Security of Uplink NOMA via Energy Harvesting Jammers
Cao, K., Wang, B., Ding, H., Lv, L., Dong, R., Cheng, T., & Gong, F. (2021) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.3023277 · Cited by 371
Deepfake detection using deep learning methods: A systematic and comprehensive review
Heidari, A., Jafari Navimipour, N., Dag, H., & Unal, M. (2024) · WIREs Data Mining and Knowledge Discovery · doi:10.1002/widm.1520 · Cited by 349 · cited 1×
“This frequently disseminated fake information and spread statue-harming info by impersonating politicians [16].”
Self-citationCross-Modal Causal Relational Reasoning for Event-Level Visual Question Answering
Liu, Y., Li, G., & Lin, L. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3284038 · Cited by 348
DeepFake Detection Based on Discrepancies Between Faces and Their Context
Nirkin, Y., Wolf, L., Keller, Y., & Hassner, T. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3093446 · Cited by 312
Multi-Region Probabilistic Histograms for Robust and Scalable Identity Inference
Sanderson, C., & Lovell, B. C. (2009) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-01793-3_21 · Cited by 301
Robust Online Tensor Completion for IoT Streaming Data Recovery
Liu, C., Wu, T., Li, Z., Ma, T., & Huang, J. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3165076 · Cited by 281
Research works printed but not linked: 3
- Darbandi M. Kalman filtering for estimation and prediction servers with lower traffic loads for transferring high-level processes in cloud computing. Published by HCTL Int J Technol Innov Res (ISSN: 2321–1814). 2017;23(1):10–20.
- Darbandi M, et al. Involving Kalman filter technique for increasing the reliability and efficiency of cloud computing. in Proceedings of the International Conference on Scientific Computing (CSC). 2012. The Steering Committee of The World Congress in Computer Science, Computer.
- Darbandi, M., Proposing new intelligent system for suggesting better service providers in cloud computing based on Kalman filtering. Published by HCTL Int J Technol Innov Res (ISSN: 2321–1814). 2017;24(1):1–9.
Cited by
Amiri, Z., Heidari, A., Zavvar, M. H., Navimipour, N. J., & Esmaeilpour, M. (2024) · Transactions on Emerging Telecommunications Technologies · doi:10.1002/ett.4969 · Cited by 129
Self-citationRazzaq, K., & Shah, M. (2025) · Computers · doi:10.3390/computers14030093 · Cited by 119
Shukla, S., Rajkumar, S., Sinha, A., Esha, M., Elango, K., & Sampath, V. (2025) · Scientific Reports · doi:10.1038/s41598-025-95858-2 · Cited by 148
ENVQA: Improving Visual Question Answering model by enriching the visual feature
Chowdhury, S., & Soni, B. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109948 · Cited by 65
Albogami, N. (2025) · Scientific Reports · doi:10.1038/s41598-025-88163-5 · Cited by 85
Metadata
- Authors
- Heidari, A., Navimipour, N. J., Dag, H., Talebi, S., & Unal, M.
- Affiliations
- Haliç University (TR)
- Kadir Has University (TR)
- The University of Texas at San Antonio (US)
- Corresponding author
- Heidari, Arash
- Navimipour, Nima Jafari
- Venue
- Cognitive Computation
- Published
- 2024-01-26
- Volume
- 16
- Issue
- 3
- Pages
- 1073-1091
- ISSN
- 1866-9956 (ISSN-L)
- 1866-9964
- OpenAlex
- W4391262065
- Cited by
- 164
- References
- 69
- Citations by year
- 2026: 38 · 2025: 95 · 2024: 31
- FWCI
- 22.84
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Cognitive Computation (publishedVersion, cc-by)
- Funders
- Kadir Has University
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Keywords
- deepfake detection, blockchain, federated learning, capsule networks, convolutional neural networks, data heterogeneity, data normalization, transfer learning, media authenticity, deepfake videos
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-28, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-09-28 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A Novel Blockchain-Based Deepfake Detection Method Using Federated and Deep Learning Models. https://app.cochat.ai/papers/10.1007/s12559-024-10255-7
- The paper
- Heidari, A., Navimipour, N. J., Dag, H., Talebi, S., & Unal, M. (2024). A Novel Blockchain-Based Deepfake Detection Method Using Federated and Deep Learning Models. Cognitive Computation. https://doi.org/10.1007/s12559-024-10255-7
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.