Methods and systems that determine heartbeat rate from video recordings
US12322205B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Mar 21, 2022 |
| Grant date | Jun 3, 2025 |
| Priority date | — |
| Expiry date | Oct 2, 2043 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30201
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
The current document is directed to computationally efficient, robust, and accurate methods and systems that determine the heartbeat rate of human subjects, patients, and/or participants monitored by a video-recording-capable device or system by analyzing video recordings of human subjects, patients, and/or participants, either in real time or following video recording. The currently disclosed methods and systems employ face-recognition methods to identify adaptive regions of interest corresponding to facial-skin areas. Spatial averages of the intensities of light components with different frequencies reflected from the identified adaptive regions of interest are extracted from the video recording as intensity signals. The intensity signals are then analyzed to extract detrended, band-pass-filtered oscillating light-intensity signals that are processed by spectral and peak-finding methods that each produces a determined heartbeat rate for each of multiple time intervals.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.