Privacy-Preserving Localization and Social Distance Monitoring with Low-Resolution Thermal Imaging and Deep Learning

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

This study introduces a novel approach to leverage low-power, low-resolution infrared sensors for detailed people tracking in manufacturing settings. We curated a dataset including a diverse range of interactions labeled for multiple-person localization and social distance violation tasks. Our methodology uses a combination of convolutional and recurrent neural networks to interpret spatiotemporal data. We demonstrate the capability of the novel image segmentation approach for human localization where we achieve 97.5 percent image-level accuracy. Also, we highlight the importance of interpolation and convolutional kernel selection for social distance tasks where we achieve 91 percent macro-averaged accuracy in 4 class scenarios.
Original languageEnglish
JournalProcedia CIRP
Volume130
Pages (from-to)355-361
Number of pages7
ISSN2212-8271
DOIs
Publication statusPublished - 01.01.2024
Event57th CIRP Conference on Manufacturing Systems - CIRP CMS 2024: Speeding up manufacturing - Universität Minho, Póvoa de Varzim , Portugal
Duration: 29.05.202431.05.2024
Conference number: 57
https://www.cirpcms2024.org/

Bibliographical note

Publisher Copyright:
© 2024 The Authors.

    Research areas

  • Deep Learning, Infrared Sensors, Convolutional Neural Networks, Facility Layout Planning, Multiple Object Localization
  • Engineering

Recently viewed

Publications

  1. Gluing life together. Computer simulation in the life sciences
  2. Using an adaptive memory strategy to improve a multistart heuristic for sequencing by hybridization
  3. Putting adaptive planning into practice: A meta-analysis of current applications
  4. Towards a Model for Building Trust and Acceptance of Artificial Intelligence Aided Medical Assessment Systems
  5. De-Anonymizing Anonymous
  6. Doing everything you can, but not (yet) getting it right
  7. Influences of yttrium content on microstructure and mechanical properties of as-cast Mg–Ca–Y–Zr alloys
  8. Compression behavior of typical silicone rubbers for soft robotics applications at elevated temperatures
  9. A Sampling Framework for Uncertainty in Individual Environmental Decisions
  10. Temporal discrimination as a function of marker duration
  11. Gaussian trajectories in motion control for camless engines
  12. Towards greener and sustainable ionic liquids using naturally occurring and nature-inspired pyridinium structures
  13. Fermentative utilization of coffee mucilage using Bacillus coagulans and investigation of down-stream processing of fermentation broth for optically pure L(+)-lactic acid production
  14. Living Labs for Product Circularity: Learnings from the ‘Innovation Network aiming at Sustainable Smartphones’
  15. Simulation-based Investigation of Energy Flexibility in the Optimization of Hinterland Drainage
  16. Digital Workplace Transformation Triggers a Shift in the HR Identity
  17. Where Tasks, Technology, and Textbooks Meet: An Exploratory Analysis of English Language Teachers’ Perceived Affordances of an Intelligent Language Tutoring System
  18. Do members of disadvantaged groups explain group status with group stereotypes?
  19. The First 50 Contributions to the Data Observer Series - An Overview
  20. Directives in EIL peer feedback
  21. Eulerian and Lagrangian perspectives on turbulent superstructures in Rayleigh-Bénard convection