University of Illinois Urbana-Champaign

Heart Rate Detection from Face Videos Using Frequency Estimation

Wang, Le

This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.

Permalink

https://hdl.handle.net/2142/55505

Description

Title
Heart Rate Detection from Face Videos Using Frequency Estimation
Author(s)
Wang, Le
Contributor(s)
Ahuja, Narendra
Issue Date
2014-05
Keyword(s)
  • Signal Processing
  • Non-Contact Heart Rate Detection
Abstract
Heart rate is a critical vital sign of physical condition in medical diagnosis and chronic disease monitoring. There are emerging needs for non-contact, low-cost, long-term and accessible cardiac pulse estimation for health monitoring and emotion assessment. Studies confirmed resting heart rate as an independent predictor of cardiovascular mortality and suggested the potential role of heart rate in future cardiovascular guidance. Moreover, both theoretical and empirical rationale support that heart rate viability helps with understanding of emotion in social and psychopathological processes. In our research, we explore a robust video-based contact-free heart rate measurement by exploring the relationship between cardiac pulse and video signals on the face. We analyze the small color variation and motion signals in the face video as a scaled and shifted version of the cardiac pulse. We empirically validate this scaled and shifted relation by a cross-correlation analysis among regions of a face. In light of the analysis, we present a spatio-temporal method to detect the heart rate from the observed video signals. Based on the frequency spectrum from the 2D spatio-temporal analysis, we estimate the cardiac pulse frequency range and use peak detection in the reconstructed wave to measure heart rate. Our method is promising on synthesized signals, and future work is required to validate the proposed method so that it could be robust to various challenging video conditions.
Type of Resource
text
Language
en
Permalink
http://hdl.handle.net/2142/55505

Owning Collections

Senior Theses - Electrical and Computer Engineering PRIMARY
The best of ECE undergraduate research