Low-Pass Image Filtering to Achieve Adversarial Robustness

Ziyadinov, V.; Tereshonok, M.

Авторы: Ziyadinov Vadim, Tereshonok Maxim
Журнал: Sensors
Том: 23
Номер: 22
Год издания: 2023
Издательство: MDPI
Местоположение издательства: Basel, Switzerland
Первая страница: 9032
Последняя страница: 9032
DOI: 10.3390/s23229032
Аннотация: In this paper, we continue the research cycle on the properties of convolutional neural network-based image recognition systems and ways to improve noise immunity and robustness. Currently, a popular research area related to artificial neural networks is adversarial attacks. The adversarial attacks on the image are not highly perceptible to the human eye, and they also drastically reduce the neural network’s accuracy. Image perception by a machine is highly dependent on the propagation of high frequency distortions throughout the network. At the same time, a human efficiently ignores high-frequency distortions, perceiving the shape of objects as a whole. We propose a technique to reduce the influence of high-frequency noise on the CNNs. We show that low-pass image filtering can improve the image recognition accuracy in the presence of high-frequency distortions in particular, caused by adversarial attacks. This technique is resource efficient and easy to implement. The proposed technique makes it possible to measure up the logic of an artificial neural network to that of a human, for whom high-frequency distortions are not decisive in object recognition.
Добавил в систему: Терешонок Максим Валерьевич

	ИСТИНА	Войти в систему Регистрация
	ИСТИНА ИНХС РАН
	Главная Поиск Статистика О проекте Помощь

ИСТИНА

ИСТИНА ИНХС РАН

Low-Pass Image Filtering to Achieve Adversarial Robustnessстатья

Прикрепленные файлы