Action Detection and Anomaly Analysis Visual System using Deep Learning for Robots in Pandemic Situation

Chia Ling Chung, Ding Bang Chen, Hooman Samani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In this paper a visual system equipped with state of the art in Deep Learning is presented which could be employed in the robotics platforms for the pandemic situation where human-human contact needs to be limited in order to perform various detection and anomaly analysis tasks. The developed detection and anomaly analysis system deals with human and environmental hazards and disasters especially for pandemic prevention. The system could detect whether the person of interest is wearing mask or not, social distancing is followed, person or environment are in normal condition for example if a window is open to keep ventilation in a closed environment. This research is a part of our project to develop a specific robot for pandemics to use advanced in artificial intelligence to make systems which could keep us safe and healthy.

Original languageEnglish
Title of host publication2020 International Automatic Control Conference, CACS 2020
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
ISBN (Electronic)9781728171982
DOIs
Publication statusPublished - 4 Nov 2020
Event2020 International Automatic Control Conference, CACS 2020 - Hsinchu, Taiwan, Province of China
Duration: 4 Nov 20207 Nov 2020

Publication series

Name2020 International Automatic Control Conference, CACS 2020

Conference

Conference2020 International Automatic Control Conference, CACS 2020
Country/TerritoryTaiwan, Province of China
CityHsinchu
Period4/11/207/11/20

Keywords

  • Anomaly Analysis
  • Coronavirus
  • Deep Learning
  • Detection
  • Pandemic
  • Robot
  • YOLO

Fingerprint

Dive into the research topics of 'Action Detection and Anomaly Analysis Visual System using Deep Learning for Robots in Pandemic Situation'. Together they form a unique fingerprint.

Cite this