EUVIMEDEUVIMED entdecken
Crossref

Non-contact respiratory rate estimation using dense optical flow and edge computing on raspberry Pi 5

Trường Hứa Mạnh, Tú Đoàn Mạnh, Thành Phạm Việt

Transport and Communications Science Journal · 25.12.2025 · 77 (5) · 745-759

DOI: 10.47869/tcsj.77.5.10 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transport and Communications Science Journal, online erschienen am 25.12.2025. Verfasst von Trường Hứa Mạnh und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The development of non-contact health monitoring systems is becoming increasingly essential in remote healthcare and early intervention support. In this paper, we introduce HealthCam, a respiratory rate monitoring system using a Raspberry Pi 5 combined with a 3.5-inch LCD and a webcam to capture real-time video. The HealthCam system identifies the shoulder–chest ROI using MediaPipe Pose, extracts respiratory signals through landmark displacement and optical flow, applies bandpass filtering (0.1–2.0 Hz), and uses multi-method peak detection, including adaptive thresholding, Hilbert phase, and derivative-based methods, to calculate BPM. By integrating edge computing, Firebase data synchronization to an authenticated web dashboard, and a dual-chatbot system (Typebot for surveys and Groq API for AI consultation), users can observe four charts: respiratory rate per minute, breath count over the measurement period, average respiratory rate per minute, and average breathing cycle. HealthCam supports time-based filtering (day, week, month, and 1–2 years) and user profile storage. Validation over 50 measurement sessions showed a Pearson correlation coefficient of r = 0.96 compared with the manual counting method, demonstrating stability under different conditions. The proposed solution offers advantages in low cost, ease of deployment, real-time operation, and AI integration to support early detection and personalized respiratory management, providing high practical applicability

Zitieren

APA

Mạnh, T. H., Mạnh, T. Đ., & Việt, T. P. (2025). Non-contact respiratory rate estimation using dense optical flow and edge computing on raspberry Pi 5. Transport and Communications Science Journal, 77(5), 745-759. https://doi.org/10.47869/tcsj.77.5.10

Vancouver

Mạnh TH, Mạnh TĐ, Việt TP. Non-contact respiratory rate estimation using dense optical flow and edge computing on raspberry Pi 5. Transport and Communications Science Journal. 2025;77(5):745-759. doi:10.47869/tcsj.77.5.10

BibTeX
@article{Mnh2025,
  author = {Mạnh, Trường Hứa and Mạnh, Tú Đoàn and Việt, Thành Phạm},
  title = {Non-contact respiratory rate estimation using dense optical flow and edge computing on raspberry Pi 5},
  journal = {Transport and Communications Science Journal},
  year = {2025},
  volume = {77},
  number = {5},
  pages = {745--759},
  doi = {10.47869/tcsj.77.5.10}
}
Kostenlos starten →