Selecting suitable deep learning network for face recognition
Islomov Shahboz Zokir ugliCryptology department, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, UzbekistanMardiyev Ulugbek RasulovichCryptology department, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, UzbekistanDavronova Lola UktamovnaCryptology department, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, UzbekistanKhamidov Sherzod Jaloldin ugliCryptology department, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, Tashkent, Uzbekistan
2019 International Conference on Information Science and Communications Technologies (ICISCT)conference2019en
ABI
Аннотация
In this paper is given problems in face recognition methods and algorithms, ways of improving of recognition accuracy and decreasing errors. There are false acceptance and rejection errors in face recognition and these errors are decreased by normalization method when detected occlusion faces. For increasing face recognition accuracy and speed is proposed combining of deep learning networks and optimization of filter size.
Темы
Идентификаторы
Цитирования и источники
Показатели — AkademScholar · Скоро