{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T04:45:40Z","timestamp":1780461940002,"version":"3.54.1"},"reference-count":22,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T00:00:00Z","timestamp":1749772800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>Convolutional neural networks (CNNs) have established themselves over time as a fundamental tool in the field of copy-move forgery detection due to their ability to effectively identify and analyze manipulated images. Unfortunately, they still represent a persistent challenge in digital image forensics, underlining the importance of ensuring the integrity of digital visual content. In this study, we present a systematic evaluation of the performance of a convolutional neural network (CNN) specifically designed for copy-move manipulation detection, applied to three datasets widely used in the literature in the context of digital forensics: CoMoFoD, Coverage, and CASIA v2. Our experimental analysis highlighted a significant variability of the results, with an accuracy ranging from 95.90% on CoMoFoD to 27.50% on Coverage. This inhomogeneity has been attributed to specific structural factors of the datasets used, such as the sample size, the degree of imbalance between classes, and the intrinsic complexity of the manipulations. We also investigated different regularization techniques and data augmentation strategies to understand their impact on the network performance, finding that adopting the L2 penalty and reducing the learning rate led to an accuracy increase of up to 2.5% for CASIA v2, while on CoMoFoD we recorded a much more modest impact (1.3%). Similarly, we observed that data augmentation was able to improve performance on large datasets but was ineffective on smaller ones. Our results challenge the idea of universal generalizability of CNN architectures in the context of copy-move forgery detection, highlighting instead how performance is strictly dependent on the intrinsic characteristics of the dataset under consideration. Finally, we propose a series of operational recommendations for optimizing the training process, the choice of the dataset, and the definition of robust evaluation protocols aimed at guiding the development of detection systems that are more reliable and generalizable.<\/jats:p>","DOI":"10.3390\/make7020054","type":"journal-article","created":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T09:51:24Z","timestamp":1749808284000},"page":"54","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["Dataset Dependency in CNN-Based Copy-Move Forgery Detection: A Multi-Dataset Comparative Analysis"],"prefix":"10.3390","volume":"7","author":[{"given":"Potito Valle","family":"Dell\u2019Olmo","sequence":"first","affiliation":[{"name":"Department of Theoretical and Applied Sciences, eCampus University, Via Isimbardi 10, 22060 Novedrate, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2331-6326","authenticated-orcid":false,"given":"Oleksandr","family":"Kuznetsov","sequence":"additional","affiliation":[{"name":"Department of Theoretical and Applied Sciences, eCampus University, Via Isimbardi 10, 22060 Novedrate, Italy"},{"name":"Department of Intelligent Software Systems and Technologies, School of Computer Science and Artificial Intelligence, V. N. Karazin Kharkiv National University, 4 Svobody Sq., 61022 Kharkiv, Ukraine"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8893-9244","authenticated-orcid":false,"given":"Emanuele","family":"Frontoni","sequence":"additional","affiliation":[{"name":"Department of Political Sciences, Communication and International Relations, University of Macerata, Via Crescimbeni, 30\/32, 62100 Macerata, Italy"},{"name":"Vision, Robotics and Artificial Intelligence Lab (VRAI), Via Brecce Bianche 12, 60131 Ancona, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1700-3075","authenticated-orcid":false,"given":"Marco","family":"Arnesano","sequence":"additional","affiliation":[{"name":"Department of Theoretical and Applied Sciences, eCampus University, Via Isimbardi 10, 22060 Novedrate, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3336-5853","authenticated-orcid":false,"given":"Christian","family":"Napoli","sequence":"additional","affiliation":[{"name":"Department of Computer, Control, and Management Engineering \u201cAntonio Ruberti\u201d, Sapienza University of Rome, V. Ariosto 25, 00185 Rome, Italy"},{"name":"Institute for Systems Analysis and Computer Science, Italian National Research Council, Via dei Taurini 19, 00185 Rome, Italy"},{"name":"Department of Computational Intelligence, Czestochowa University of Technology, ul Dabrowskiego 69, 42-201 Czestochowa, Poland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5300-3561","authenticated-orcid":false,"given":"Cristian","family":"Randieri","sequence":"additional","affiliation":[{"name":"Department of Theoretical and Applied Sciences, eCampus University, Via Isimbardi 10, 22060 Novedrate, Italy"},{"name":"Department of Computer, Control, and Management Engineering \u201cAntonio Ruberti\u201d, Sapienza University of Rome, V. Ariosto 25, 00185 Rome, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Huang, H.-Y., and Ciou, A.-J. (2019). Copy-Move Forgery Detection for Image Forensics Using the Superpixel Segmentation and the Helmert Transformation. EURASIP J. Image Video Process., 2019.","DOI":"10.1186\/s13640-019-0469-9"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"659","DOI":"10.1016\/j.image.2013.03.006","article-title":"Copy-Move Forgery Detection and Localization by Means of Robust Clustering with J-Linkage","volume":"28","author":"Amerini","year":"2013","journal-title":"Signal Process. Image Commun."},{"key":"ref_3","first-page":"336","article-title":"Copy-Move Forgery Detection from Printed Images","volume":"Volume 9028","author":"Amerini","year":"2014","journal-title":"Proceedings of the Media Watermarking, Security, and Forensics 2014"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Mahmood, T., Nawaz, T., Irtaza, A., Ashraf, R., Shah, M., and Mahmood, M.T. (2016). Copy-Move Forgery Detection Technique for Forensic Analysis in Digital Images. Math. Probl. Eng., 2016.","DOI":"10.1155\/2016\/8713202"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"15171","DOI":"10.1007\/s11042-022-12010-2","article-title":"A Comprehensive Evaluation Procedure for Copy-Move Forgery Detection Methods: Results from a Systematic Review","volume":"81","author":"Warif","year":"2022","journal-title":"Multimed. Tools Appl."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1016\/j.jnca.2016.09.008","article-title":"Copy-Move Forgery Detection: Survey, Challenges and Future Directions","volume":"75","author":"Warif","year":"2016","journal-title":"J. Netw. Comput. Appl."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Caldelli, R., Amerini, I., and Costanzo, A. (September, January 31). SIFT Match Removal and Keypoint Preservation through Dominant Orientation Shift. Proceedings of the 2015 23rd European Signal Processing Conference (EUSIPCO), Nice, France.","DOI":"10.1109\/EUSIPCO.2015.7362747"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Amerini, I., Barni, M., Caldelli, R., and Costanzo, A. (2013). Counter-Forensics of SIFT-Based Copy-Move Detection by Means of Keypoint Classification. J. Image Video Proc., 2013.","DOI":"10.1186\/1687-5281-2013-18"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"213","DOI":"10.1016\/j.neucom.2022.09.017","article-title":"QDL-CMFD: A Quality-Independent and Deep Learning-Based Copy-Move Image Forgery Detection Method","volume":"511","author":"Aria","year":"2022","journal-title":"Neurocomputing"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"El_Tokhy, M.S. (2023). Development of Precise Forgery Detection Algorithms in Digital Radiography Images Using Convolution Neural Network. Appl. Soft Comput., 138.","DOI":"10.1016\/j.asoc.2023.110174"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1099","DOI":"10.1109\/TIFS.2011.2129512","article-title":"A SIFT-Based Forensic Method for Copy\u2013Move Attack Detection and Transformation Recovery","volume":"6","author":"Amerini","year":"2011","journal-title":"IEEE Trans. Inf. Forensics Secur."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1016\/j.icte.2021.08.016","article-title":"An Optimized Technique for Copy\u2013Move Forgery Localization Using Statistical Features","volume":"8","author":"Babu","year":"2022","journal-title":"ICT Express"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Ayd\u0131n, Y. (2022). A New Copy-Move Forgery Detection Method Using LIOP. J. Vis. Commun. Image Represent., 89.","DOI":"10.1016\/j.jvcir.2022.103661"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"19167","DOI":"10.1007\/s11042-020-08751-7","article-title":"A Novel Deep Learning Framework for Copy-Moveforgery Detection in Images","volume":"79","author":"Elaskily","year":"2020","journal-title":"Multimed. Tools Appl."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"59783","DOI":"10.1007\/s11042-023-17964-5","article-title":"Enhancing Copy-Move Forgery Detection through a Novel CNN Architecture and Comprehensive Dataset Analysis","volume":"83","author":"Kuznetsov","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Kumar, S., Gupta, S.K., Kaur, M., and Gupta, U. (2022). VI-NET: A Hybrid Deep Convolutional Neural Network Using VGG and Inception V3 Model for Copy-Move Forgery Classification. J. Vis. Commun. Image Represent., 89.","DOI":"10.1016\/j.jvcir.2022.103644"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Wang, J., Gao, X., Nie, J., Wang, X., Huang, L., Nie, W., Jiang, M., and Wei, Z. (2024). Strong Robust Copy-Move Forgery Detection Network Based on Layer-by-Layer Decoupling Refinement. Inf. Process. Manag., 61.","DOI":"10.1016\/j.ipm.2024.103685"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Ravula, J., and Singh, N. (2025). GAN-ViT-CMFD: A Novel Framework Integrating Generative Adversarial Networks and Vision Transformers for Enhanced Copy-Move Forgery Detection and Classification with Spectral Clustering. Intell. Syst. Appl., 26.","DOI":"10.1016\/j.iswa.2025.200524"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Nazir, T., Nawaz, M., Masood, M., and Javed, A. (2022). Copy Move Forgery Detection and Segmentation Using Improved Mask Region-Based Convolution Network (RCNN). Appl. Soft Comput., 131.","DOI":"10.1016\/j.asoc.2022.109778"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Rutkowski, L., Scherer, R., Korytkowski, M., Pedrycz, W., Tadeusiewicz, R., and Zurada, J.M. (2024, January 16\u201320). Embedding Non-Distortive Cancelable Face Template Generation. Proceedings of the Artificial Intelligence and Soft Computing, Zakopane, Poland.","DOI":"10.1007\/978-3-031-84353-2"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Shaushenova, A., Kuznetsov, O., Nurpeisova, A., and Ongarbayeva, M. (2025). Implementation of Kolmogorov\u2013Arnold Networks for Efficient Image Processing in Resource-Constrained Internet of Things Devices. Technologies, 13.","DOI":"10.3390\/technologies13040155"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Rutkowski, L., Scherer, R., Korytkowski, M., Pedrycz, W., Tadeusiewicz, R., and Zurada, J.M. (2024, January 16\u201320). A Novel Approach to Cancelable Biometrics Through Deep Learning Distortion Metrics. Proceedings of the Artificial Intelligence and Soft Computing, Zakopane, Poland.","DOI":"10.1007\/978-3-031-84353-2"}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/7\/2\/54\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:51:39Z","timestamp":1760032299000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/7\/2\/54"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,13]]},"references-count":22,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["make7020054"],"URL":"https:\/\/doi.org\/10.3390\/make7020054","relation":{},"ISSN":["2504-4990"],"issn-type":[{"value":"2504-4990","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,13]]}}}