Respiratory rate estimation using nostril detection in thermal video streams

Bogdan Mocanu, Dragos Ţarǎlungǎ, Ruxandra Ţapu, Ermina Ţapu

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

Abstract

In this paper we propose a complete method for breath rate estimation using thermal video streams and computer vision algorithms. The system performs subject nose detection, interest point extraction and tracking, geometric transformation and minimum thermal region estimation. All algorithms are carefully designed in order to ensure real-time capability on a regular desktop computer. The experimental evaluation validates the proposed methodology, when compared with traditional breath rate extraction methods, returning high accuracy scores.

Original languageEnglish
Title of host publication2017 E-Health and Bioengineering Conference, EHB 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages201-204
Number of pages4
ISBN (Electronic)9781538603581
DOIs
Publication statusPublished - 28 Jul 2017
Externally publishedYes
Event6th IEEE International Conference on E-Health and Bioengineering, EHB 2017 - Sinaia, Romania
Duration: 22 Jun 201724 Jun 2017

Publication series

Name2017 E-Health and Bioengineering Conference, EHB 2017

Conference

Conference6th IEEE International Conference on E-Health and Bioengineering, EHB 2017
Country/TerritoryRomania
CitySinaia
Period22/06/1724/06/17

Keywords

  • breath rate estimation
  • nostril tracking and detection
  • thermal video stream

Fingerprint

Dive into the research topics of 'Respiratory rate estimation using nostril detection in thermal video streams'. Together they form a unique fingerprint.

Cite this