System and methods for vital sign estimation from passive thermal video
First Claim
1. A system for estimating vital signs from passive thermal video comprising:
- an infrared video receiving module for receiving an infrared video segment of a subject;
a contour segmentation module for contour segmenting the video segment to select a region of pixels representing a portion of the subject;
an alignment module for aligning the selected region of pixels across a plurality of frames of the video segment;
a signal detection module for detecting a signal from the selected region using a thermal wave propagation-based signal detection method;
a spatial filtering module for spatial filtering the signal to remove noise not related to the vital signs to be estimated;
a non-linear temporal filtering module to process a temporal signal corresponding to each aligned pixel sequence;
an clustering module for removing outliers of the signal using a pixel clustering algorithm;
a frequency selection module for selecting a frequency peak of the signal using a dominant frequency voting technique that generates a plurality of bins based on a sampling rate, each of the plurality of bins representing a possible frequency peak of the signal, votes on each of the plurality of bins and selects the frequency peak having a highest number of votes; and
an averaging module for averaging the selected frequency peak to compute an average estimation of at least one vital sign.
2 Assignments
0 Petitions
Accused Products
Abstract
A system for measuring a pulse and respiratory rate from passive thermal video includes contour segmentation and tracking, clustering of informative pixels of interests, and robust dominant frequency component estimation. Contour segmentation is used to locate a blood vessel region to measure, after which all pixels in the nearby region are aligned across frames based on the segmentation'"'"'s position, and scale in each frame. Spatial filtering is then performed to remove noise not related to heart beat and then non-linear filtering is performed on the temporal signal corresponding to each aligned pixel. The signal spectrum of each pixel is then feed to a clustering algorithm for outlier removal. Pixels in the largest cluster are then used to vote for the dominant frequency, and the median of the dominant frequency is output as the pulse rate.
-
Citations
20 Claims
-
1. A system for estimating vital signs from passive thermal video comprising:
-
an infrared video receiving module for receiving an infrared video segment of a subject; a contour segmentation module for contour segmenting the video segment to select a region of pixels representing a portion of the subject; an alignment module for aligning the selected region of pixels across a plurality of frames of the video segment; a signal detection module for detecting a signal from the selected region using a thermal wave propagation-based signal detection method; a spatial filtering module for spatial filtering the signal to remove noise not related to the vital signs to be estimated; a non-linear temporal filtering module to process a temporal signal corresponding to each aligned pixel sequence; an clustering module for removing outliers of the signal using a pixel clustering algorithm; a frequency selection module for selecting a frequency peak of the signal using a dominant frequency voting technique that generates a plurality of bins based on a sampling rate, each of the plurality of bins representing a possible frequency peak of the signal, votes on each of the plurality of bins and selects the frequency peak having a highest number of votes; and an averaging module for averaging the selected frequency peak to compute an average estimation of at least one vital sign. - View Dependent Claims (2, 3, 4, 5, 6, 19, 20)
-
-
7. A method for estimating vital signs from a passive thermal video comprising the steps of
receiving an infrared video segment of a subject; -
contour segmenting the video segment to select a region of pixels representing a portion of the subject; aligning the selected region of pixels across a plurality of frames of the video segment; detecting a signal from the selected region using a thermal wave propagation-based signal detection method; spatial filtering the signal to remove noise not related to the vital signs to be estimated; performing non-linear filtering on the signal corresponding to each aligned pixel of the selected region; removing outliers of the signal using a pixel clustering algorithm; selecting a frequency peak of the signal using a dominant frequency voting technique that generates a plurality of bins based on a sampling rate, each of the plurality of bins representing a possible frequency peak of the signal, votes on each of the plurality of bins and selects the frequency peak having a highest number of votes; and averaging the selected frequency peak to compute an average estimation of at least one vital sign. - View Dependent Claims (8, 9, 10, 11, 12)
-
-
13. A computer program product embodied on a non-transitory computer readable medium for estimating vital signs from passive thermal video comprising computer program code for:
-
receiving an infrared video segment of a subject; contour segmenting the video segment to select a region of pixels representing a portion of the subject; aligning the selected region of pixels across a plurality of frames of the video segment; detecting a signal from the selected region using a thermal wave propagation-based signal detection method; spatial filtering the signal to remove noise not related to the vital signs to be estimated; performing non-linear filtering on the signal corresponding to each aligned pixel of the selected region; removing outliers of the signal using a pixel clustering algorithm; selecting a frequency peak of the signal using a dominant frequency voting technique that generates a plurality of bins based on a sampling rate, each of the plurality of bins representing a possible frequency peak of the signal, votes on each of the plurality of bins and selects the frequency peak having a highest number of votes; and averaging the selected frequency peak to compute an average estimation of at least one vital sign. - View Dependent Claims (14, 15, 16, 17, 18)
-
Specification