Low-Pass Image Filtering to Achieve Adversarial Robustness

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.

Automated summary

This paper continues the research on the properties of convolutional neural network-based image recognition systems and explores ways to enhance the systems' immunity to noise and robustness. It focuses on adversarial attacks, which are not easily perceptible to human eyes but greatly reduce the accuracy of the neural networks. The paper proposes a technique that utilizes low-pass image filtering to mitigate the influence of high-frequency distortions caused by adversarial attacks, thereby improving image recognition accuracy. The technique is resource-efficient and easy to implement, bridging the gap between artificial neural networks and human recognition logic.