{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T09:00:00Z","timestamp":1774602000573,"version":"3.50.1"},"reference-count":45,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T00:00:00Z","timestamp":1774569600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"The authors received no funding for this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>With the rapid development of artificial intelligence generation technology, the boundary between artificial intelligence (AI)-generated images and real images is becoming increasingly blurred, posing serious challenges to the credibility and authenticity of digital content. Addressing the insufficient generalization of existing AI-generated image detection methods in complex scenarios, this research proposes a Diffusion-Cross Attention Transformer (DCAT) framework for image authenticity verification. This framework innovatively combines diffusion model feature extractors and cross-attention vision transformers (ViT) to achieve fine-grained capture of image microscopic noise distribution and semantic relationships. Large-scale experimental validation was conducted on the GenImage dataset. The model demonstrated excellent performance in various degradation environments, with area under the receiver operating characteristic curve (AUC) remaining stable from 0.910 under no degradation conditions to 0.775 in extreme degradation environments, significantly outperforming traditional methods. The core contributions of this research include proposing a multi-scale noise analysis feature extraction method, constructing a cross-attention semantic association detection mechanism, and theoretically deepening the mathematical characterization of distribution differences between generated models and real images. This innovative approach not only provides key technological breakthroughs but also offers important technical support for maintaining the authenticity of digital content ecosystems, holding significant scientific and practical value for the field of artificial intelligence image generation and detection.<\/jats:p>","DOI":"10.7717\/peerj-cs.3655","type":"journal-article","created":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T08:13:16Z","timestamp":1774599196000},"page":"e3655","source":"Crossref","is-referenced-by-count":0,"title":["Cross-generator image authenticity verification: multi-modal feature fusion and distribution difference analysis"],"prefix":"10.7717","volume":"12","author":[{"given":"Yi","family":"Li","sequence":"first","affiliation":[{"name":"College of Humanities and Social Sciences, Huazhong Agricultural University, Wuhan, China"},{"name":"Faculty of Art, Sustainability and Creative Industry, Sultan Idris Education University, Perak, Malaysia"},{"name":"School of Design, Hebei Academy of Fine Arts, Hebei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ida Puteri","family":"Mahsan","sequence":"additional","affiliation":[{"name":"Faculty of Art, Sustainability and Creative Industry, Sultan Idris Education University, Perak, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wenjun","family":"Yang","sequence":"additional","affiliation":[{"name":"Faculty of Computing and Meta-Technology, Sultan Idris Education University, Perak, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanjun","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Humanities and Social Sciences, Huazhong Agricultural University, Wuhan, China"},{"name":"Faculty of Art, Sustainability and Creative Industry, Sultan Idris Education University, Perak, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2026,3,27]]},"reference":[{"key":"10.7717\/peerj-cs.3655\/ref-1","doi-asserted-by":"publisher","first-page":"147772\u2013147783","DOI":"10.1109\/access.2024.3466614","article-title":"Detection of AI-generated images from various generators using gated expert convolutional neural network","volume":"12","author":"Ahmad Fattah Saskoro","year":"2024","journal-title":"IEEE Access"},{"issue":"2","key":"10.7717\/peerj-cs.3655\/ref-2","doi-asserted-by":"publisher","first-page":"222","DOI":"10.5505\/pajes.2023.94395","article-title":"Copy-move forgery detection using EOA, DWT and DCT","volume":"30","author":"Amiri","year":"2024","journal-title":"Pamukkale University Journal of Engineering Sciences-Pamukkale Universitesi Muhendislik Bilimleri Dergisi"},{"key":"10.7717\/peerj-cs.3655\/ref-3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ojsp.2023.3337714","article-title":"Synthbuster: towards detection of diffusion model generated images","volume":"5","author":"Bammey","year":"2024","journal-title":"IEEE Open Journal of Signal Processing"},{"issue":"10","key":"10.7717\/peerj-cs.3655\/ref-4","doi-asserted-by":"publisher","first-page":"199","DOI":"10.3390\/jimaging9100199","article-title":"AI vs. AI: can AI detect AI-generated images?","volume":"9","author":"Baraheem","year":"2023","journal-title":"Journal of Imaging"},{"key":"10.7717\/peerj-cs.3655\/ref-5","doi-asserted-by":"publisher","DOI":"10.1109\/InC460750.2024.10649158","article-title":"Detecting AI-generated images with CNN and interpretation using explainable AI","author":"Bharathi Mohan","year":"2024"},{"issue":"3","key":"10.7717\/peerj-cs.3655\/ref-6","doi-asserted-by":"publisher","first-page":"2212","DOI":"10.1109\/TPAMI.2024.3522305","article-title":"RenAIssance: a survey into AI text-to-image generation in the era of large model","volume":"47","author":"Bie","year":"2025","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.7717\/peerj-cs.3655\/ref-7","first-page":"157","article-title":"AI generated art: latent diffusion-based style and detection","volume-title":"Advances in Computational Intelligence Systems, UKCI 2023, volume 1453 of Advances in Intelligent Systems and Computing","author":"Bird","year":"2024"},{"key":"10.7717\/peerj-cs.3655\/ref-8","doi-asserted-by":"publisher","first-page":"15642","DOI":"10.1109\/access.2024.3356122","article-title":"Cifake: image classification and explainable identification of AI-generated synthetic images","volume":"12","author":"Bird","year":"2024","journal-title":"IEEE Access"},{"issue":"7","key":"10.7717\/peerj-cs.3655\/ref-9","doi-asserted-by":"publisher","first-page":"1372","DOI":"10.14569\/ijacsa.2024.01507133","article-title":"IPD-Net: detecting ai-generated images via inter-patch dependencies","volume":"15","author":"Chen","year":"2024","journal-title":"International Journal of Advanced Computer Science and Applications"},{"issue":"16","key":"10.7717\/peerj-cs.3655\/ref-10","doi-asserted-by":"publisher","first-page":"17871","DOI":"10.1609\/aaai.v38i16.29741","article-title":"How to protect copyright data in optimization of large language models?","volume":"38","author":"Chu","year":"2024","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"10.7717\/peerj-cs.3655\/ref-11","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP49357.2023.10095167","article-title":"On the detection of synthetic images generated by diffusion models","author":"Corvi","year":"2023"},{"key":"10.7717\/peerj-cs.3655\/ref-12","first-page":"4356","article-title":"Raising the bar of AI-generated image detection with clip","author":"Cozzolino","year":"2024"},{"key":"10.7717\/peerj-cs.3655\/ref-13","first-page":"410","article-title":"Cultural relevance index: measuring cultural relevance in AI-generated images","author":"ELsharif","year":"2024"},{"key":"10.7717\/peerj-cs.3655\/ref-14","first-page":"382","article-title":"Online detection of AI-generated images","author":"Epstein","year":"2023"},{"key":"10.7717\/peerj-cs.3655\/ref-15","first-page":"4822","article-title":"Organic or diffused: can we distinguish human art from AI-generated images?","author":"Ha","year":"2024"},{"issue":"1","key":"10.7717\/peerj-cs.3655\/ref-16","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1080\/20004214.2024.2340787","article-title":"Photorealism versus photography: AI-generated depiction in the age of visual disinformation","volume":"16","author":"Hausken","year":"2024","journal-title":"Journal of Aesthetics & Culture"},{"issue":"5","key":"10.7717\/peerj-cs.3655\/ref-17","doi-asserted-by":"publisher","first-page":"341","DOI":"10.1016\/j.inffus.2023.02.026","article-title":"Multiscale structural feature transform for multi-modal image matching","volume":"95","author":"Hu","year":"2023","journal-title":"Information Fusion"},{"key":"10.7717\/peerj-cs.3655\/ref-18","first-page":"1168","article-title":"Evading watermark based detection of AI-generated content","author":"Jiang","year":"2023"},{"issue":"6","key":"10.7717\/peerj-cs.3655\/ref-19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3633284","article-title":"Detecting post editing of multimedia images using transfer learning and fine tuning","volume":"20","author":"Jonker","year":"2024","journal-title":"ACM Transactions on Multimedia Computing Communications and Applications"},{"key":"10.7717\/peerj-cs.3655\/ref-20","article-title":"Leveraging representations from intermediate encoder-blocks for synthetic image detection","author":"Koutlis","year":"2024"},{"issue":"1","key":"10.7717\/peerj-cs.3655\/ref-21","doi-asserted-by":"publisher","first-page":"110","DOI":"10.1007\/s40319-023-01419-3","article-title":"Copyright law and the lifecycle of machine learning models","volume":"55","author":"Kretschmer","year":"2024","journal-title":"IIC-International Review of Intellectual Property and Competition Law"},{"issue":"3","key":"10.7717\/peerj-cs.3655\/ref-22","doi-asserted-by":"publisher","first-page":"1575","DOI":"10.3390\/ai5030076","article-title":"Detection of AI-generated synthetic images with a lightweight CNN","volume":"5","author":"Ladevic","year":"2024","journal-title":"AI"},{"key":"10.7717\/peerj-cs.3655\/ref-23","doi-asserted-by":"publisher","first-page":"3644","DOI":"10.1109\/access.2024.3522759","article-title":"Advanced detection of AI-generated images through vision transformers","volume":"13","author":"Lamichhane","year":"2025","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.3655\/ref-24","first-page":"3855","article-title":"Masksim: detection of synthetic images by masked spectrum similarity analysis","author":"Li","year":"2024"},{"issue":"6","key":"10.7717\/peerj-cs.3655\/ref-25","doi-asserted-by":"publisher","first-page":"17069","DOI":"10.1007\/s11042-023-16217-9","article-title":"AMFF-Net: adaptive multi-modal feature fusion network for image classification","volume":"83","author":"Liu","year":"2024","journal-title":"Multimedia Tools and Applications"},{"issue":"5","key":"10.7717\/peerj-cs.3655\/ref-26","doi-asserted-by":"publisher","first-page":"979","DOI":"10.1002\/ase.2336","article-title":"Evaluating AI-powered text-to-image generators for anatomical illustration: a comparative study","volume":"17","author":"Noel","year":"2024","journal-title":"Anatomical Sciences Education"},{"key":"10.7717\/peerj-cs.3655\/ref-27","first-page":"24480","article-title":"Towards universal fake image detectors that generalize across generative models","author":"Ojha","year":"2023"},{"key":"10.7717\/peerj-cs.3655\/ref-28","first-page":"299","article-title":"Can you spot the AI-generated images? Distinguishing fake images using signal detection theory","author":"Park","year":"2024"},{"issue":"11","key":"10.7717\/peerj-cs.3655\/ref-29","doi-asserted-by":"publisher","first-page":"62609","DOI":"10.1109\/access.2024.3394250","article-title":"Performance comparison and visualization of AI-generated-image detection methods","volume":"12","author":"Park","year":"2024","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.3655\/ref-30","first-page":"534","article-title":"Authenticating AI-generated social media images using frequency domain analysis","author":"Poredi","year":"2024"},{"issue":"4","key":"10.7717\/peerj-cs.3655\/ref-31","doi-asserted-by":"publisher","first-page":"1375","DOI":"10.1111\/1556-4029.70069","article-title":"Exploring machine learning approaches for efficient image forgery detection","volume":"70","author":"Radhakrishnan","year":"2025","journal-title":"Journal of Forensic Sciences"},{"issue":"14","key":"10.7717\/peerj-cs.3655\/ref-32","doi-asserted-by":"publisher","first-page":"3134","DOI":"10.3390\/math11143134","article-title":"Review of image forensic techniques based on deep learning","volume":"11","author":"Shi","year":"2023","journal-title":"Mathematics"},{"key":"10.7717\/peerj-cs.3655\/ref-33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tim.2024.3451570","article-title":"DAE-Net: dual attention mechanism and edge supervision network for image manipulation detection and localization","volume":"73","author":"Shi","year":"2024","journal-title":"IEEE Transactions on Instrumentation and Measurement"},{"issue":"10","key":"10.7717\/peerj-cs.3655\/ref-34","doi-asserted-by":"publisher","first-page":"4746","DOI":"10.1007\/s11263-024-02116-5","article-title":"Watcher: wavelet-guided texture-content hierarchical relation learning for deepfake detection","volume":"132","author":"Wang","year":"2024","journal-title":"International Journal of Computer Vision"},{"key":"10.7717\/peerj-cs.3655\/ref-35","first-page":"7278","article-title":"Dynamic graph learning with content guided spatial-frequency relation reasoning for deepfake detection","author":"Wang","year":"2023a"},{"issue":"12","key":"10.7717\/peerj-cs.3655\/ref-36","doi-asserted-by":"publisher","first-page":"2593","DOI":"10.3390\/math11122593","article-title":"A frequency attention-based dual-stream network for image inpainting forensics","volume":"11","author":"Wang","year":"2023b","journal-title":"Mathematics"},{"issue":"14","key":"10.7717\/peerj-cs.3655\/ref-37","doi-asserted-by":"publisher","first-page":"3192","DOI":"10.3390\/electronics12143192","article-title":"A multi-path inpainting forensics network based on frequency attention and boundary guidance","volume":"12","author":"Wang","year":"2023c","journal-title":"Electronics"},{"key":"10.7717\/peerj-cs.3655\/ref-38","first-page":"1463","article-title":"AI-generated image detection using a cross-attention enhanced dual-stream network","author":"Xi","year":"2023"},{"issue":"4","key":"10.7717\/peerj-cs.3655\/ref-39","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1016\/j.patrec.2023.10.021","article-title":"Exposing fake images generated by text-to-image diffusion models","volume":"176","author":"Xu","year":"2023","journal-title":"Pattern recognition Letters"},{"key":"10.7717\/peerj-cs.3655\/ref-40","doi-asserted-by":"publisher","first-page":"109924","DOI":"10.1016\/j.compag.2025.109924","article-title":"Effective multi-species weed detection in complex wheat fields using multi-modal and multi-view image fusion","volume":"230","author":"Xu","year":"2025a","journal-title":"Computers and Electronics in Agriculture"},{"issue":"86","key":"10.7717\/peerj-cs.3655\/ref-41","doi-asserted-by":"publisher","first-page":"226","DOI":"10.1109\/lsp.2024.3511421","article-title":"FAMSeC: a few-shot-sample-based general AI-generated image detection method","volume":"32","author":"Xu","year":"2025b","journal-title":"IEEE Signal Processing Letters"},{"key":"10.7717\/peerj-cs.3655\/ref-42","first-page":"11964","article-title":"Editguard: versatile image watermarking for tamper localization and copyright protection","author":"Zhang","year":"2024"},{"issue":"9","key":"10.7717\/peerj-cs.3655\/ref-43","doi-asserted-by":"publisher","first-page":"1877","DOI":"10.3390\/electronics14091877","article-title":"Multi-scale edge-guided image forgery detection via improved self-supervision and self-adversarial training","volume":"14","author":"Zhang","year":"2025","journal-title":"Electronics"},{"issue":"4","key":"10.7717\/peerj-cs.3655\/ref-44","doi-asserted-by":"publisher","first-page":"129683","DOI":"10.1016\/j.neucom.2025.129683","article-title":"A spatio-frequency cross fusion model for deepfake detection and segmentation","volume":"628","author":"Zheng","year":"2025","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.3655\/ref-45","article-title":"Genimage: a million-scale benchmark for detecting AI-generated image","author":"Zhu","year":"2023"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3655.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3655.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3655.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3655.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T08:13:19Z","timestamp":1774599199000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3655"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,27]]},"references-count":45,"alternative-id":["10.7717\/peerj-cs.3655"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3655","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,27]]},"article-number":"e3655"}}