Персона: Борзунов, Георгий Иванович
Загружается...
Email Address
Birth Date
Научные группы
Организационные подразделения
Организационная единица
Институт интеллектуальных кибернетических систем
Цель ИИКС и стратегия развития - это подготовка кадров, способных противостоять современным угрозам и вызовам, обладающих знаниями и компетенциями в области кибернетики, информационной и финансовой безопасности для решения задач разработки базового программного обеспечения, повышения защищенности критически важных информационных систем и противодействия отмыванию денег, полученных преступным путем, и финансированию терроризма.
Статус
Фамилия
Борзунов
Имя
Георгий Иванович
Имя
3 results
Результаты поиска
Теперь показываю 1 - 3 из 3
- ПубликацияОткрытый доступOptimization of machine learning algorithm of emotion recognition in terms of human facial expressions(2020) Ivanova, E.; Borzunov, G.; Борзунов, Георгий Иванович© 2020 The Authors. Published by Elsevier B.V.This work is devoted to the optimization of the recognition method of seven basic emotions (joy, sadness, fear, anger, surprise, disgust and neutral) in terms of the expressions of the human face. The existing approaches of the emotion recognition systems construction was analyzed based on human facial expressions, and focused on the advantages of using scheems based on neural networks. We proposed a method of constructing an emotion recognition system based on a neural network, which includes an optimized algorithm for generating training and test samples, as well as determining the rational number of layers of the neural network.
- ПубликацияТолько метаданныеImproving the Security of the Facial Biometrics System Using the Liveness Detection Module(2020) Ivanova, E.; Borzunov, G.; Борзунов, Георгий Иванович© 2020, Springer Nature Switzerland AG.Biometric methods are of considerable value when used alone or in combination with other identity verification technologies. Two-dimensional facial recognition approaches provide low cost and convenient recognition system due to convenience and ease of use. Rapid face image substitution is one of the main problems in 2D face area. Biometric systems can be attacked by fakes such as images of people’s faces, masks and videos that are easily accessible from social networks. The typical disadvantage of survivability detection in consumer-grade methods is a significant disadvantage and limits the value of device-built biometric authentication in smartphones and tablets. The work is devoted to the study of methods for verifying the belonging of a biometric sample to a living person. The relevance of the work is due to the expansion of the use of biometric authentication systems and the need to protect the biometric identification and authentication processes from hacking attempts using photographs or video. For the experimental evaluation of the complex application of the studied methods, a prototype of a multi-module system for testing faces using neural networks and heuristic algorithms was developed.
- ПубликацияТолько метаданныеAnalysis and optimization of the packing tree search algorithm for the knapsack problem(2019) Slonkina, I.; Kupriyashin, M.; Borzunov, G.; Куприяшин, Михаил Андреевич; Борзунов, Георгий Иванович© 2019 IEEE The Packing Tree Search algorithm proved to be one of the best exact algorithms for the Knapsack Problem in cryptography: while free from the memory-related limitations of the Horowitz-Sahni list merging algorithm, it still offers a significant performance boost if compared to the Exhaustive Search algorithm. The algorithm also seems to be scalable, albeit the parallel computation efficiency may be as low as 50-60%. In this paper, we consider improvements to the base algorithm to reduce its complexity. In particular, we study the dynamic packing weight calculation, the replacement of stack-based and linearization-based tree traversal techniques with faster step-based traversal routines and substitution of the original problem instance with the conjugate in case it reduces the complexity. In conclusion, we analyze the impact of these improvements on the acceleration and efficiency of parallel computation.