Skip to main navigation Skip to search Skip to main content

Risk Analysis of One-Pixel Image Defects in Safety-Critical Deep Neural Networks

  • Krystian Radlak
  • , Adam Popowicz
  • , Michal Szczepankiewicz
  • , Pawel Zawistowski
  • Warsaw University of Technology
  • UL Solutions
  • NVIDIA

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Deep neural networks (DNNs) are widely considered essential for developing perception systems in autonomous applications. These models are often vulnerable to small perturbations in input data, even if the changes appear negligible to a human observer. This vulnerability introduces an additional risk of failure in safety-critical systems during normal operation. Unfortunately, there is currently no quantitative risk analysis addressing such image defects. In contrast, this work examines the risk that one-pixel defects may occur naturally within image data, and evaluates how frequently such seemingly minor defects can lead to incorrect decisions by neural networks. Extensive experiments reveal that the number of impactful image defects may be relatively high, depending on both the DNN architecture and the dataset used. These findings establish that image defects require significant attention and it might not be sufficient to argue for an acceptable level of safety based solely on the low probability of occurrence these defects.

Publication series

NameLecture Notes in Computer Science
Volume15955 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceCo-Design of Communication, Computing and Control in Cyber-Physical Systems, CoC3CPS 2025, 20th Workshop on Dependable Smart Embedded and Cyber-Physical Systems and Systems-of-Systems, DECSoS 2025, 12th International Workshop on Next Generation of System Assurance Approaches for Critical Systems, SASSUR 2025, 4th International Workshop on Safety and Security Interaction, SENSEI 2025, 2nd International Workshop on Safety/Reliability/Trustworthiness of Intelligent Transportation Systems, SRToITS 2025 and 8th International Workshop on Artificial Intelligence Safety Engineering, WAISE 2025 held in conjunction with the 44th International Conference on Computer Safety, Reliability, and Security, SAFECOMP 2025
Country/TerritorySweden
CityStockholm
Period9/09/259/09/25

Keywords

  • deep neural networks
  • image defects
  • safety
  • security

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Risk Analysis of One-Pixel Image Defects in Safety-Critical Deep Neural Networks'. Together they form a unique fingerprint.

Cite this