Semi-supervised deep learning approach to break common CAPTCHAs

Simple item page
dc.contributor.authorBoštík, Ondřejcs
dc.contributor.authorHorák, Karelcs
dc.contributor.authorKratochvíla, Lukášcs
dc.contributor.authorZemčík, Tomášcs
dc.contributor.authorBilík, Šimoncs
dc.coverage.issue20cs
dc.coverage.volume33cs
dc.date.issued2021-04-12cs
dc.description.abstractManual data annotation is a time consuming activity. A novel strategy for automatic training of the CAPTCHA breaking system with no manual dataset creation is presented in this paper. We demonstrate the feasibility of the attack against a text-based CAPTCHA scheme utilizing similar network infrastructure used for Denial of Service attacks. The main goal of our research is to present a possible vulnerability in CAPTCHA systems when combining the brute-force attack with transfer learning. The classification step utilizes a simple convolutional neural network with 15 layers. Training stage uses automatically prepared dataset created without any human intervention and transfer learning for fine-tuning the deep neural network classifier. The designed system for breaking text-based CAPTCHAs achieved 80% classification accuracy after 6 fine-tuning steps for a 5 digit text-based CAPTCHA system. The results presented in this paper suggest, that even the simple attack with a large number of attacking computers can be an effective alternative to current CAPTCHA breaking systems.en
dc.formattextcs
dc.format.extent13333-13343cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationNEURAL COMPUTING & APPLICATIONS. 2021, vol. 33, issue 20, p. 13333-13343.en
dc.identifier.doi10.1007/s00521-021-05957-0cs
dc.identifier.issn0941-0643cs
dc.identifier.orcid0000-0002-7856-2084cs
dc.identifier.orcid0000-0002-2280-3029cs
dc.identifier.orcid0000-0001-8425-323Xcs
dc.identifier.orcid0000-0003-4363-4313cs
dc.identifier.orcid0000-0001-8797-7700cs
dc.identifier.other170906cs
dc.identifier.researcheridJEP-7714-2023cs
dc.identifier.scopus57222421244cs
dc.identifier.urihttp://hdl.handle.net/11012/203005
dc.language.isoencs
dc.publisherSpringercs
dc.relation.ispartofNEURAL COMPUTING & APPLICATIONScs
dc.relation.urihttps://link.springer.com/article/10.1007%2Fs00521-021-05957-0cs
dc.rights(C) Springercs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/0941-0643/cs
dc.subjectCAPTCHAen
dc.subjectSemi-supervised learningen
dc.subjectConvolutional Neural Networksen
dc.titleSemi-supervised deep learning approach to break common CAPTCHAsen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionacceptedVersionen
sync.item.dbidVAV-170906en
sync.item.dbtypeVAVen
sync.item.insts2025.02.03 15:39:30en
sync.item.modts2025.01.17 15:21:58en
thesis.grantorVysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. Ústav automatizace a měřicí technikycs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
05957_accepted.pdf
Size:
371.06 KB
Format:
Adobe Portable Document Format
Description:
05957_accepted.pdf
Download
Collections
Ústav automatizace a měřicí techniky