Evaluating the Vulnerability of YOLOv5 to Adversarial Attacks for Enhanced Cybersecurity in MASS

Abstract

The development of artificial intelligence (AI) technologies, such as machine learning algorithms, computer vision systems, and sensors, has allowed maritime autonomous surface ships (MASS) to navigate, detect and avoid obstacles, and make real-time decisions based on their environment. Despite the benefits of AI in MASS, its potential security threats must be considered. An adversarial attack is a security threat that involves manipulating the training data of a model to compromise its accuracy and reliability. This study focuses on security threats faced by a deep neural network-based object classification algorithm, particularly you only look once version 5 (YOLOv5), which is a model used for object classification. We performed transfer learning on YOLOv5 and tested various adversarial attack methods. We conducted experiments using four types of adversarial attack methods and parameter changes to determine the attacks that could be detrimental to YOLOv5. Through this study, we aim to raise awareness of the vulnerability of AI algorithms for object detection to adversarial attacks and emphasize the need for efforts to overcome them; these efforts can contribute to safe navigation in MASS.

Comprehensive Analysis of Maritime Cybersecurity Landscape Based on the NIST CSF v2.0

“…Similarly, Charitos et al [40] addressed the susceptibility of merchant shipping to cyberattacks, particularly focusing on the threats impacting embedded technologies, and underscored the need for efficient threat management. Complementing this perspective, Lee et al [41] presented an evaluation of security threats to maritime autonomous surface ships, further illuminating AI algorithm vulnerabilities. Bolbot et al [42] suggested a methodology for cyber risk assessment, particularly for autonomous ships, emphasizing the need for implementing measures such as firewalls and intrusion detection systems.…”