0525022 20240103224144.020200616235959.9 10.2352/ISSN.2470-1173.2020.6.IRIACV-052 5 s. E cav_un_epca*05252362470-1173SpringfieldSociety for Imaging Science and Technology2020 super resolution generative adversial networks optical character recognition cav_un_auth*0332672 Bílková Zuzana UTIA-B Zpracování obrazové informace Department of Image Processing ZOI ZOI Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0322281 Hradiš M. CZ http://library.utia.cas.cz/separaty/2020/ZOI/bilkova-0525022.pdf SVV-2017-260452 GA MŠk CZ cav_un_auth*0393278 1583117 GA UK CZ cav_un_auth*0393033 AP1701 AV ČR CZ cav_un_auth*0349658 We present a novel method for super-resolution (SR) of license plate images based on an end-to-end convolutional neural networks (CNN) combining generative adversial networks (GANs) and optical character recognition (OCR). License plate SR systems play an important role in number of security applications such as improvement of road safety, traffic monitoring or surveillance. The specific task requires not only realistic-looking reconstructed images but it also needs to preserve the text information. Standard CNN SR and GANs fail to accomplish this requirment. The incorporation of the OCR pipeline into the method also allows training of the network without the need of ground truth high resolution data which enables easy training on real data with all the real image degradations including compression. cav_un_auth*0392692 IS&T International Symposium on Electronic Imaging 2020 20200126 20200130 Burlingame, California US JC 10000 10200 10201 2021 2 PR RVO:67985556 http://hdl.handle.net/11104/0309418 1 cav_un_auth*0300003 Vysoké učení technické v Brně, Fakulta informačních technologií VUT FIT S 052 2020 SCOPUS PUBMED WOS cav_un_epca*0525236 Electronic Imaging, Intelligent Robotics and Industrial Applications using Computer Vision 2020 2470-1173 Springfield Society for Imaging Science and Technology 2020