Russian Federation
The article made a study of methods and computer vision algorithms based on deep convolutional and recurrent neural networks. Image processing for the purpose of their recognition is one of the central and practically important tasks in creation of artificial intelligence systems. Face detection and recognition in images is a classic task in computer vision and the most modem trend in intelligent security and access control systems, in user authorization, in organizing video conferencing, in robotics and biometrics. Considered evolution of approaches and algorithms to problems solving of processing, recognition and localization of faces in images using Viola-Jones object detection algorithm, convolutional and recurrent neural networks. Technologies for the restoration of a three-dimensional image of a face from an initial 2D image by neural methods are considered. The results of computer experiments based on a convolutional and recurrent neural network for face detection, age determination, face emotional state and mask detection in real-time video stream are implemented in Matlab R2020a. Practical application of the obtained results can be used in distant education in identifying the personality of cadets and students.
pattern recognition, computer vision, deep machine learning, Viola-Jones object detection, Haar-like features, convolutional and recurrent neural networks, emotional artificial intelligence
1. Studenikin D.E., Bovan S.D., Hekert E.V., Modina M.A. Ispol'zovanie neyronnyh setey dlya organizacii vizual'nogo nablyudeniya // Morskie intellektual'nye tehnologii,- 2019 - № 4-3 (46). - S. 91-95.
2. Brilyuk D.V., Starovoytov V.V. Raspoznavanie cheloveka po izobrazheniyu lica neyrosetevymi metodami. - Minsk, 2002. - 54 s.
3. LeCun Y. Deep Learning Hardware: Past, Present, and Future. IEEE International Solid-State Circuits Conference-(ISSCC), 2019. - 12-19 pp.
4. Klette P. Komp'yuternoe zrenie. Teoriya i algoritmy / per. s angl. A.A. Slinkin. - M.: DMK Press, 2019. - 506 s.
5. Polkovnikova N.A., Polkovnikov A.K. Sistema podderzhki prinyatiya resheniy dlya vybora rezhima i prognozirovaniya otkazov glavnogo sudovogo dvigatelya // Ekspluataciya morskogo transporta.-2019,-№3(92).-S. 170-180.
6. Polkovnikova N.A. Neyrosetevye sistemy avtomaticheskogo raspoznavaniya morskih ob'ektov // Ekspluataciya morskogo transporta. - 2020,- №1(94). - S. 207-219.
7. Belyh E. A. Obuchenie kaskadov Haara // Vestnik Syktyvkarskogo universiteta. Ser. 1: Matematika. Mehanika. Informatika,- 2017-Vyp. 1 (22).-S. 41-53.
8. Chang Buy Thi Thu, Hoang Fan Ngok, Spicyn V.G. Raspoznavanie lic na osnove primeneniya metoda Violy-Dzhonsa, veyvlet-preobrazovaniya i metoda glavnyh komponent // Izvestiya Tomskogo politehnicheskogo universiteta-2012.- t. 320,- № 5,- S. 54-59.
9. Sheshkus A.V., Matalov D.P., Arlazarov V.V., Nikolaev D.P. Issledovanie kompozicii algoritmov komp'yuternogo zreniya, baziruyuschihsya na mashinnom obuchenii, dlya resheniya zadachi lokalizacii i klassifikacii ob'ektov // Trudy Instituta sistemnogo analiza Rossiyskoy akademii nauk- 2019-T. 69-№ 1. - S. 29-36.
10. Schroff F., Kalenichenko D., Philbin J. FaceNet: A Unified Embedding for Face Recognition and Clustering https://arxiv.org/abs/1503.03832 (Data obrascheniya 05.08.2020)
11. Slipchenko O.O., Klimenko S.V., Stepanova A.Yu. Ispol'zovanie apparata neyrosetey pri raspoznavanii lica // Aktual'ni problemi avtomatizacii ta informaciynih tehnologiy.-2017.- tom 21.- S. 143-151
12. Polkovnikova N.A. Sistema podderzhki prinyatiya resheniy na osnove svertochnyh neyronnyh setey po raspoznavaniyu ob'ektov na izobrazheniyah // Materialy III mezhdunarodnogo nauchnotehnicheskogo foruma «Sovremennye tehnologii v nauke i obrazovanii» (4-6 marta 2020, Ryazan'). - Ryazan': Ryazanskiy gosudarstvennyy radiotehnicheskiy universitet imeni V.F. Utkina, t.5.-s. 108-112.
13. Recent advances on memetic algorithms and its applications in image processing. Editors: Hemanth D.J., Kumar B.V., Karpagam Manavalan G.R. -Springer, 2020. - 199 p.
14. Leskovec J., Rajaraman A., Ullman J.D. Mining of Massive Datasets. - Cambridge University Press, 3rd edition, 2019. - 583 p.
15. Redmon J., Farhadi A. YOL09000: Better, Faster, Stronger. Proceedings of the IEEE conference on computer vision and pattern recognition, 2017. -7263-7271 pp.
16. LeCun Y., Bengio Y., Hinton G. Deep learning. Nature, vol. 521 (7553), 2015.-436-444 pp.
17. Krizhevsky A., Sutskever I., Hinton G.E. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 2012. - 1097-1105 pp.
