{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T17:33:57Z","timestamp":1779384837427,"version":"3.53.1"},"reference-count":24,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,4,9]],"date-time":"2024-04-09T00:00:00Z","timestamp":1712620800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:p>Rail surface defects present a significant safety concern in railway operations. However, the scarcity of data poses challenges for employing deep learning in defect detection. This study proposes an enhanced ACGAN augmentation method to address these issues. Residual blocks mitigate vanishing gradient problems, while a spectral norm regularization-constrained discriminator improves stability and image quality. Substituting the generator\u2019s deconvolution layer with upsampling and convolution operations enhances computational efficiency. A gradient penalty mechanism based on regret values addresses gradient abnormality concerns. Experimental validation demonstrates superior image clarity and classification accuracy compared to ACGAN, with a 17.6% reduction in FID value. MNIST dataset experiments verify the model\u2019s generalization ability. This approach offers practical value for real-world applications.<\/jats:p>","DOI":"10.3389\/fnbot.2024.1397369","type":"journal-article","created":{"date-parts":[[2024,4,9]],"date-time":"2024-04-09T10:23:32Z","timestamp":1712658212000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Rail surface defect data enhancement method based on improved ACGAN"],"prefix":"10.3389","volume":"18","author":[{"given":"He","family":"Zhendong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gao","family":"Xiangyang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liu","family":"Zhiyuan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"An","family":"Xiaoyu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zheng","family":"Anping","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2024,4,9]]},"reference":[{"key":"ref1","first-page":"511","article-title":"Survey of few-shot image classification research journal of Frontiers of computer","volume":"17","author":"An","year":"2023","journal-title":"Sci. Technol."},{"key":"ref2","first-page":"214","article-title":"Wasserstein generative adversarial networks","author":"Arjovsky","year":"2017","journal-title":"Int. Conf. Machine Learn."},{"key":"ref3","first-page":"31841","article-title":"Get 3d: a generative model of high quality 3d textured shapes learned from images","volume":"35","author":"Gao","year":"2022","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1145\/3422622","article-title":"Generative adversarial networks","volume":"63","author":"Goodfellow","year":"2020","journal-title":"Commun. ACM"},{"key":"ref5","first-page":"1","article-title":"Improved training of wasserstein gans","volume":"30","author":"Gulrajani","year":"2017","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref6","first-page":"3383","article-title":"Fault self-diagnosis of structural vibration monitoring sensor and monitoring data recovery based on CNN and DCGAN","volume":"19","author":"Guo","year":"2022","journal-title":"J. Railway Sci. Eng."},{"key":"ref7","first-page":"1","article-title":"Gans trained by a two time-scale update rule converge to a local Nash equilibrium","volume":"30","author":"Heusel","year":"2017","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref8","first-page":"1102","article-title":"Object detector with residual learning and multi-scale feature enhancement","volume":"17","author":"Jia","year":"2023","journal-title":"J. Front. Comput. Sci. Technol."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"ref10","first-page":"1904","article-title":"Low-light enhancement method for light field images by fusing multi-scale features","volume":"17","author":"Li","year":"2023","journal-title":"J. Front. Comput. Sci. Technol."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"101552","DOI":"10.1016\/j.aei.2022.101552","article-title":"Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework","volume":"52","author":"Li","year":"2022","journal-title":"Adv. Eng. Inform."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1016\/j.neunet.2020.10.016","article-title":"Adversarial symmetric GANs: bridging adversarial samples and adversarial networks","volume":"133","author":"Liu","year":"2021","journal-title":"Neural Netw."},{"key":"ref13","first-page":"2794","article-title":"Least squares generative adversarial networks","author":"Mao","year":"2017","journal-title":"Proc. IEEE Int. Conf. Comput. Vision"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2021.3125987","article-title":"An adaptive pyramid graph and variation residual-based anomaly detection network for rail surface defects","volume":"70","author":"Niu","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref15","first-page":"2642","article-title":"Conditional image synthesis with auxiliary classifier gans","author":"Odena","year":"2017","journal-title":"Int. Conf. Mach. Learn."},{"key":"ref16","first-page":"2431","article-title":"LSTM-based distributed conditional generative adversarial network for data-driven 5G-enabled maritime UAV communications","volume":"24","author":"Rasheed","year":"2023","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"5641","DOI":"10.1109\/TIP.2020.2985545","article-title":"CGAN-TM: a novel domain-to-domain transferring method for person re-identification","volume":"29","author":"Tang","year":"2020","journal-title":"IEEE Trans. Image Process."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2020.3026801","article-title":"Deep learning-based generic automatic surface defect inspection (ASDI) with pixelwise segmentation","volume":"70","author":"Wu","year":"2020","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"656","DOI":"10.1109\/TIM.2018.2853958","article-title":"A coarse-to-fine model for rail surface defect detection","volume":"68","author":"Yu","year":"2018","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref20","first-page":"15","article-title":"A review of deep contrast learning","volume":"49","author":"Zhang","year":"2023","journal-title":"Acta Automat. Sin."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1438","DOI":"10.1109\/TWC.2021.3103971","article-title":"Distributed conditional generative adversarial networks (GANs) for data-driven millimeter wave communications in UAV networks","volume":"21","author":"Zhang","year":"2021","journal-title":"IEEE Trans. Wirel. Commun."},{"key":"ref22","first-page":"7354","article-title":"Self-attention generative adversarial networks","author":"Zhang","year":"2019","journal-title":"Int. Conf. Mach. Learn."},{"key":"ref23","first-page":"11","article-title":"Review of nondestructive testing and evaluation techniques for rail defects","volume":"40","author":"Zhang","year":"2019","journal-title":"Chin. J. Sci. Instrum."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1016\/j.neucom.2022.05.021","article-title":"A novel MAS-GAN-based data synthesis method for object surface defect detection","volume":"499","author":"Zhang","year":"2022","journal-title":"Neurocomputing"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1397369\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,4,9]],"date-time":"2024-04-09T10:23:43Z","timestamp":1712658223000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1397369\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,9]]},"references-count":24,"alternative-id":["10.3389\/fnbot.2024.1397369"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2024.1397369","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,4,9]]},"article-number":"1397369"}}