Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Robust Camera-based Method for Breath Rate Measurement

Alexey Protopopov

TL;DR

This paper tackles non-contact breath rate monitoring from video by addressing motion artifacts that degrade accuracy in prior camera-based methods. It introduces a robust pipeline that dynamically segments the chest region, emphasizes vertical chest motion via optical flow, and extracts BR from a refined signal using multi-stage filtering and peak detection. On 14 video recordings (8 male, 6 female; total ~2.5 hours) captured with a standard webcam, the method achieved MAE of 0.57 rpm and relative error under 4%, outperforming several prior approaches. The approach is hardware-light and suitable for remote monitoring, though real-time performance awaits faster segmentation hardware and shorter processing delays. The work also highlights potential applications in clinical appointments and driver drowsiness detection.

Abstract

Proliferation of cheap and accessible cameras makes it possible to measure a subject's breath rate from video footage alone. Recent works on this topic have proposed a variety of approaches for accurately measuring human breath rate, however they are either tested in near-ideal conditions, or produce results that are not sufficiently accurate. The present study proposes a more robust method to measure breath rate in humans with minimal hardware requirements using a combination of mathematical transforms with a relative deviation from the ground truth of less than 5%. The method was tested on videos taken from 14 volunteers with a total duration of over 2 hours 30 minutes. The obtained results were compared to reference data and the average mean absolute error was found to be at 0.57 respirations per minute, which is noticeably better than the results from previous works. The breath rate measurement method proposed in the present article is more resistant to distortions caused by subject movement and thus allows one to remotely measure the subject's breath rate without any significant limitations on the subject's behavior.

A Robust Camera-based Method for Breath Rate Measurement

TL;DR

This paper tackles non-contact breath rate monitoring from video by addressing motion artifacts that degrade accuracy in prior camera-based methods. It introduces a robust pipeline that dynamically segments the chest region, emphasizes vertical chest motion via optical flow, and extracts BR from a refined signal using multi-stage filtering and peak detection. On 14 video recordings (8 male, 6 female; total ~2.5 hours) captured with a standard webcam, the method achieved MAE of 0.57 rpm and relative error under 4%, outperforming several prior approaches. The approach is hardware-light and suitable for remote monitoring, though real-time performance awaits faster segmentation hardware and shorter processing delays. The work also highlights potential applications in clinical appointments and driver drowsiness detection.

Abstract

Proliferation of cheap and accessible cameras makes it possible to measure a subject's breath rate from video footage alone. Recent works on this topic have proposed a variety of approaches for accurately measuring human breath rate, however they are either tested in near-ideal conditions, or produce results that are not sufficiently accurate. The present study proposes a more robust method to measure breath rate in humans with minimal hardware requirements using a combination of mathematical transforms with a relative deviation from the ground truth of less than 5%. The method was tested on videos taken from 14 volunteers with a total duration of over 2 hours 30 minutes. The obtained results were compared to reference data and the average mean absolute error was found to be at 0.57 respirations per minute, which is noticeably better than the results from previous works. The breath rate measurement method proposed in the present article is more resistant to distortions caused by subject movement and thus allows one to remotely measure the subject's breath rate without any significant limitations on the subject's behavior.

Paper Structure

This paper contains 8 sections, 4 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Subject segmentation
  4. Motion calculation
  5. Signal processing
  6. Results and discussion
  7. Conclusion
  8. Acknowledgements

Figures (4)

  • Figure 1: Direction filter for movement vectors. Vectors that deviate too much from the vertical axis were discarded.
  • Figure 2: The result of signal normalization. The filtered signal is drawn in black, the polylines are drawn in green, the resulting normalized signal — in red.
  • Figure 3: Comparison of a signal from the video to a reference signal.
  • Figure 4: Comparison of the video BR to the reference BR. Black — BR acquired from the video, red — BR according to the upper breath sensor, orange — BR according to the lower breath sensor.