{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,11]],"date-time":"2025-11-11T15:49:25Z","timestamp":1762876165752,"version":"3.41.0"},"reference-count":76,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2020,4,17]],"date-time":"2020-04-17T00:00:00Z","timestamp":1587081600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100012659","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61972314, 61672427, 61872294"],"award-info":[{"award-number":["61972314, 61672427, 61872294"]}],"id":[{"id":"10.13039\/501100012659","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Ant Financial Science Funds for Security Research"},{"name":"International Cooperation Project of Shaanxi Province","award":["2019KW-009"],"award-info":[{"award-number":["2019KW-009"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Priv. Secur."],"published-print":{"date-parts":[[2020,5,31]]},"abstract":"<jats:p>Text-based CAPTCHAs remains a popular scheme for distinguishing between a legitimate human user and an automated program. This article presents a novel genetic text captcha solver based on the generative adversarial network. As a departure from prior text captcha solvers that require a labor-intensive and time-consuming process to construct, our scheme needs significantly fewer real captchas but yields better performance in solving captchas. Our approach works by first learning a synthesizer to automatically generate synthetic captchas to construct a base solver. It then improves and fine-tunes the base solver using a small number of labeled real captchas. As a result, our attack requires only a small set of manually labeled captchas, which reduces the cost of launching an attack on a captcha scheme. We evaluate our scheme by applying it to 33 captcha schemes, of which 11 are currently used by 32 of the top-50 popular websites. Experimental results demonstrate that our scheme significantly outperforms four prior captcha solvers and can solve captcha schemes where others fail. As a countermeasure, we propose to add imperceptible perturbations onto a captcha image. We demonstrate that our countermeasure can greatly reduce the success rate of the attack.<\/jats:p>","DOI":"10.1145\/3378446","type":"journal-article","created":{"date-parts":[[2020,5,4]],"date-time":"2020-05-04T07:04:40Z","timestamp":1588575880000},"page":"1-29","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":19,"title":["Using Generative Adversarial Networks to Break and Protect Text Captchas"],"prefix":"10.1145","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2074-4253","authenticated-orcid":false,"given":"Guixin","family":"Ye","sequence":"first","affiliation":[{"name":"Northwest University, China"}]},{"given":"Zhanyong","family":"Tang","sequence":"additional","affiliation":[{"name":"Northwest University, China"}]},{"given":"Dingyi","family":"Fang","sequence":"additional","affiliation":[{"name":"Northwest University, China"}]},{"given":"Zhanxing","family":"Zhu","sequence":"additional","affiliation":[{"name":"Peking University, China"}]},{"given":"Yansong","family":"Feng","sequence":"additional","affiliation":[{"name":"Peking University, China"}]},{"given":"Pengfei","family":"Xu","sequence":"additional","affiliation":[{"name":"Northwest University, China"}]},{"given":"Xiaojiang","family":"Chen","sequence":"additional","affiliation":[{"name":"Northwest University, China"}]},{"given":"Jungong","family":"Han","sequence":"additional","affiliation":[{"name":"University of Warwick, United Kingdom"}]},{"given":"Zheng","family":"Wang","sequence":"additional","affiliation":[{"name":"University of Leeds, United Kingdom"}]}],"member":"320","published-online":{"date-parts":[[2020,4,17]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1145\/2991079.2991083"},{"volume-title":"Proceedings of the International Conference on Machine Learning. 214--223","year":"2017","author":"Arjovsky Martin","key":"e_1_2_1_2_1"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/11909033_9"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1137\/040603371"},{"volume-title":"Proceedings of the ACM Symposium on Information, Computer and Communications Security. 16--25","author":"Barreno Marco","key":"e_1_2_1_5_1"},{"volume-title":"Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 1829--1838","author":"Jeffrey","key":"e_1_2_1_6_1"},{"key":"e_1_2_1_7_1","unstructured":"Elie Bursztein. 2012. How we Broke the NuCaptcha Video Scheme and What we Proposed to Fix it. Retrieved from https:\/\/elie.net\/blog\/security\/how-we-broke-the-nucaptcha-video-scheme-and-what-we-propose-to-fix-it.  Elie Bursztein. 2012. How we Broke the NuCaptcha Video Scheme and What we Proposed to Fix it. Retrieved from https:\/\/elie.net\/blog\/security\/how-we-broke-the-nucaptcha-video-scheme-and-what-we-propose-to-fix-it."},{"volume-title":"Proceedings of the USENIX Workshop on Offensive Technologies (WOOT\u201914)","author":"Bursztein Elie","key":"e_1_2_1_8_1"},{"volume-title":"Proceedings of the Usenix Conference on Offensive Technologies. 8--8.","year":"2009","author":"Bursztein Elie","key":"e_1_2_1_9_1"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1145\/2046707.2046724"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/2556288.2557322"},{"volume-title":"Proceedings of the Conference on Email 8 Anti-Spam.","year":"2005","author":"Chellapilla Kumar","key":"e_1_2_1_12_1"},{"volume-title":"Proceedings of the International Conference on Information Security. Springer, 268--279","author":"Chew Monica","key":"e_1_2_1_13_1"},{"volume-title":"Proceedings of the ACM Conference on Computer and Communications Security (CCS\u201907)","year":"2007","author":"Elson Jeremy","key":"e_1_2_1_14_1"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT.2004.1365067"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2017.2682704"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/2508859.2516732"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2016.23154"},{"volume-title":"An Evolutionary Study of Dynamic Cognitive Game CAPTCHAs: Automated Attacks and Defenses. Dissertations Theses Gradworks","author":"Gao Song","key":"e_1_2_1_19_1"},{"key":"e_1_2_1_20_1","doi-asserted-by":"crossref","unstructured":"Dileep George Wolfgang Lehrach Ken Kansky Miguel L\u00e1zaro-Gredilla Christopher Laan Bhaskara Marthi Xinghua Lou Zhaoshi Meng Yi Liu Huayan Wang etal 2017. A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs. Science 358 6368 (2017) eaag2612.  Dileep George Wolfgang Lehrach Ken Kansky Miguel L\u00e1zaro-Gredilla Christopher Laan Bhaskara Marthi Xinghua Lou Zhaoshi Meng Yi Liu Huayan Wang et al. 2017. A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs. Science 358 6368 (2017) eaag2612.","DOI":"10.1126\/science.aag2612"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2005.06.044"},{"key":"e_1_2_1_22_1","first-page":"535","article-title":"Machine learning attacks against the Asirra CAPTCHA","volume":"2008","author":"Golle Philippe","year":"2008","journal-title":"Comput. Commun. Secur."},{"volume-title":"Proceedings of the International Conference on Learning Representations (ICLR\u201914)","year":"2014","author":"Goodfellow Ian J.","key":"e_1_2_1_23_1"},{"key":"e_1_2_1_24_1","first-page":"2672","article-title":"Generative adversarial networks","volume":"3","author":"Goodfellow Ian J.","year":"2014","journal-title":"Adv. Neural Info. Process. Syst."},{"volume-title":"Proceedings of the International Conference on Machine Learning (ICML\u201915)","year":"2015","author":"Goodfellow Ian J.","key":"e_1_2_1_25_1"},{"key":"e_1_2_1_26_1","unstructured":"Ian J. Goodfellow David Warde-Farley Mehdi Mirza Aaron Courville and Yoshua Bengio. 2013. Maxout networks. arXiv preprint arXiv:1302.4389.  Ian J. Goodfellow David Warde-Farley Mehdi Mirza Aaron Courville and Yoshua Bengio. 2013. Maxout networks. arXiv preprint arXiv:1302.4389."},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/1526709.1526822"},{"volume-title":"Proceedings of the IEEE Computer Society Conferene on Computer Vision and Pattern Recognition.","year":"2003","author":"Greg Mori","key":"e_1_2_1_28_1"},{"volume-title":"Mask R-CNN. In Proceedings of the IEEE International Conference on Computer Vision (ICCV\u201917)","year":"2017","author":"He Kaiming","key":"e_1_2_1_29_1"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_1_31_1","doi-asserted-by":"crossref","unstructured":"Robert Hecht-Nielsen. 1989. Theory of the Backpropagation Neural Network. Harcourt Brace 8 Co. 593\u2013605.  Robert Hecht-Nielsen. 1989. Theory of the Backpropagation Neural Network. Harcourt Brace 8 Co. 593\u2013605.","DOI":"10.1109\/IJCNN.1989.118638"},{"key":"e_1_2_1_32_1","first-page":"1185","article-title":"Side-channel attack on labeling CAPTCHAs","volume":"0908","author":"Hernandezcastro Carlos Javier","year":"2009","journal-title":"Comput. Sci. ArXiv Preprint ArXiv"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/MIC.2011.112"},{"key":"e_1_2_1_34_1","doi-asserted-by":"crossref","unstructured":"Phillip Isola. 2017. Pix2Pix: Image-to-Image Translation with COnditional Adversarial Networks. Retrieved from https:\/\/github.com\/phillipi\/pix2pix.  Phillip Isola. 2017. Pix2Pix: Image-to-Image Translation with COnditional Adversarial Networks. Retrieved from https:\/\/github.com\/phillipi\/pix2pix.","DOI":"10.1109\/CVPR.2017.632"},{"volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1125--1134","author":"Isola Phillip","key":"e_1_2_1_35_1"},{"volume-title":"Strong captcha guidelines v1. 2 [J]. 10","year":"2010","author":"Wilkins J.","key":"e_1_2_1_36_1"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/INFCOM.2013.6567061"},{"volume-title":"Proceedings of the International Conference on Learning Representations.","author":"Diederik","key":"e_1_2_1_38_1"},{"volume-title":"Proceedings of the NDSS Workshop on Usable Security.","author":"Krol Kat","key":"e_1_2_1_39_1"},{"volume-title":"Proceedings of the European Conference on Computer Vision. 47--54","author":"Lea Colin","key":"e_1_2_1_40_1"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1230"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/TDSC.2018.2874243"},{"key":"e_1_2_1_44_1","unstructured":"Min Lin Qiang Chen and Shuicheng Yan. 2013. Network in network. arXiv preprint arXiv:1312.4400.  Min Lin Qiang Chen and Shuicheng Yan. 2013. Network in network. arXiv preprint arXiv:1312.4400."},{"key":"e_1_2_1_45_1","unstructured":"Elaine K. McEwan. 2008. Root Words Roots and Affixes. Retrieved from http:\/\/www.readingrockets.org\/article\/root-words-roots-and-affixes.  Elaine K. McEwan. 2008. Root Words Roots and Affixes. Retrieved from http:\/\/www.readingrockets.org\/article\/root-words-roots-and-affixes."},{"volume-title":"Proceedings of the International Conference on Machine Learning and Applications. 67--73","year":"2014","author":"Meutzner Hendrik","key":"e_1_2_1_46_1"},{"volume-title":"Proceedings of the International Conference on Learning Representations (Poster).","year":"2015","author":"Miyato Takeru","key":"e_1_2_1_47_1"},{"key":"e_1_2_1_48_1","first-page":"5602","article-title":"Playing atari with deep reinforcement learning","volume":"1312","author":"Mnih Volodymyr","year":"2013","journal-title":"ArXiv Preprint ArXiv"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/2590296.2590298"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.3233\/JCS-16847"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2017.2718479"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2009.191"},{"key":"e_1_2_1_53_1","doi-asserted-by":"crossref","unstructured":"Ishai Rosenberg Asaf Shabtai Lior Rokach and Yuval Elovici. 2017. Generic black-box end-to-end attack against RNNs and other API calls based malware classifiers. arXiv preprint arXiv:1707.05970.  Ishai Rosenberg Asaf Shabtai Lior Rokach and Yuval Elovici. 2017. Generic black-box end-to-end attack against RNNs and other API calls based malware classifiers. arXiv preprint arXiv:1707.05970.","DOI":"10.1007\/978-3-030-00470-5_23"},{"key":"e_1_2_1_54_1","unstructured":"Neil J. Rubenking. 2013. Are You a Human. Retrieved from https:\/\/www.areyouahuman.com.  Neil J. Rubenking. 2013. Are You a Human. Retrieved from https:\/\/www.areyouahuman.com."},{"key":"e_1_2_1_56_1","unstructured":"NuData Security. 2010. NuCaptcha. Retrieved from www.nucaptcha.com.  NuData Security. 2010. NuCaptcha. Retrieved from www.nucaptcha.com."},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2016.2635643"},{"key":"e_1_2_1_58_1","unstructured":"Chenghui Shi Xiaogang Xu Shouling Ji Kai Bu Jianhai Chen Raheem A. Beyah and Ting Wang. 2019. Adversarial CAPTCHAs. Retrieved from http:\/\/arxiv.org\/abs\/1901.01107  Chenghui Shi Xiaogang Xu Shouling Ji Kai Bu Jianhai Chen Raheem A. Beyah and Ting Wang. 2019. Adversarial CAPTCHAs. Retrieved from http:\/\/arxiv.org\/abs\/1901.01107"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.241"},{"volume-title":"In Proceedings of the International Conference on Learning Representations.","year":"2014","author":"Simonyan Karen","key":"e_1_2_1_60_1"},{"volume-title":"Proceedings of the IEEE European Symposium on Security and Privacy. 388--403","author":"Sivakorn Suphannee","key":"e_1_2_1_61_1"},{"volume-title":"Proceedings of the German Conference on Pattern Recognition Workshop.","year":"2015","author":"Stark Fabian","key":"e_1_2_1_62_1"},{"key":"e_1_2_1_63_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_1_64_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"volume-title":"Proceedings of the International Conference on Learning Representations (ICLR).","year":"2014","author":"Szegedy Christian","key":"e_1_2_1_65_1"},{"volume-title":"Proceedings of the Conference on Neural Information Processing Systems. 1625--1632","year":"2008","author":"Tam Jennifer","key":"e_1_2_1_66_1"},{"volume-title":"CAPTCHA: Using Hard AI Problems for Security","year":"2003","author":"Ahn Luis Von","key":"e_1_2_1_67_1"},{"key":"e_1_2_1_68_1","doi-asserted-by":"publisher","DOI":"10.1145\/966389.966390"},{"volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR\u201919)","author":"Xie Cihang","key":"e_1_2_1_69_1"},{"key":"e_1_2_1_70_1","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2016.23115"},{"key":"e_1_2_1_71_1","doi-asserted-by":"publisher","DOI":"10.1109\/TDSC.2013.52"},{"key":"e_1_2_1_72_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACSAC.2007.47"},{"key":"e_1_2_1_73_1","doi-asserted-by":"publisher","DOI":"10.1145\/1455770.1455839"},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.1145\/3243734.3243754"},{"volume-title":"How transferable are features in deep neural networks? In Advances in Neural Information Processing Systems","author":"Yosinski Jason","key":"e_1_2_1_75_1"},{"volume-title":"Proceedings of the Thirty-First Association for the Advancement of Artificial Intelligence. 2852\u20132858","year":"2016","author":"Yu Lantao","key":"e_1_2_1_76_1"},{"key":"e_1_2_1_77_1","unstructured":"Jun-Yan Zhu Taesung Park Phillip Isola and Alexei A Efros. 2017. Unpaired image-to-image translation using cycle-consistent adversarial networks. arXiv preprint arXiv:1703.10593.  Jun-Yan Zhu Taesung Park Phillip Isola and Alexei A Efros. 2017. Unpaired image-to-image translation using cycle-consistent adversarial networks. arXiv preprint arXiv:1703.10593."}],"container-title":["ACM Transactions on Privacy and Security"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3378446","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3378446","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T23:45:04Z","timestamp":1750203904000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3378446"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,4,17]]},"references-count":76,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2020,5,31]]}},"alternative-id":["10.1145\/3378446"],"URL":"https:\/\/doi.org\/10.1145\/3378446","relation":{},"ISSN":["2471-2566","2471-2574"],"issn-type":[{"type":"print","value":"2471-2566"},{"type":"electronic","value":"2471-2574"}],"subject":[],"published":{"date-parts":[[2020,4,17]]},"assertion":[{"value":"2019-06-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-01-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-04-17","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}