This paper presents a real-time heart rate monitoring system using video from a webcam. It extracts the heart rate from variations in facial skin color caused by blood circulation. Three signal processing methods - Fast Fourier Transform, Independent Component Analysis, and Principal Component Analysis - are applied to color channels in the video to extract the blood volume pulse. The extracted heart rate is then compared to a reference rate. Results show there is a high degree of agreement between the proposed method and the reference readings, indicating this non-contact technique has potential for use in personal healthcare and telemedicine applications.

1. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 03 | Mar 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 417
Pulse Rate Monitoring Using Image Processing
Akhilesh Kulkarni[1], Aniket Mishra [2], Mandar Mohite[3], Prof. Jignesh Patel [4]
[1], [2], [3]Student, Department of Computer Engineering, Atharva College of Engineering, Mumbai
[4]Professor, Department of Computer Engineering, Atharva College of Engineering, Mumbai
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract – Heart Rate (HR) is one of the most important
physiological features and an important indicator of
human health and therefore important to monitor. HR
monitoring often involves high costs and complex use of
sensors and sensory systems. Ongoing research over the
past decade has largely focused on offline systems that are
simpler, less expensive and more comfortable to use.
However, many unrelated systems are suitable for lab
locations in offline environments but need to be
significantly improved before they can be used in real-
time applications. This paper shows a real-time HR
monitoring system using a portable computer webcam.
Heart rate is determined by a variation of facial skin
colour caused by blood circulation. Three different signal
processing methods such as Fast Fourier Transform (FFT),
Independent Component Analysis (ICA) and Principal
Component Analysis (PCA) have been used in colour
channels in video recording and the volume pulse (BVP) is
output to the surface. regions. HR is priced and compared
with corresponding reference ratings. The results
obtained indicate that there is a high degree of agreement
between the proposed tests and the reference ratings. This
technology has great potential to improve personal health
care and telemedicine. The continuous development of the
proposed algorithm considering natural light and
movement may be very useful in many real-time programs
such as driver monitoring.
Key Words: Pulse rate monitoring, Haar cascade
classifier algorithm, OpenCV based image processing
implementation, independent component analysis
(ICA), Region of Interest (ROI)
1.INTRODUCTION
Sensing the heartbeat without actual contact utilising
signal and image handling is perhaps the most effective
way. Image and signal handling is the utilisation of PC
calculations to deal with image processing and work on
the exactness of advanced pictures/recordings. It permits
a much more extensive scope of calculations to be utilised
in input information and can stay away from issues like
sound arrangement and signal falsification during
processing. Pulse rate variations can be estimated in a
straightforward recorded face video that makes testing
simpler, less expensive. Our fundamental methodology is
to consider the timeseries of colour values at any point to
increase the variations at a given frequency band of
interest.
2. LITERATURE SURVEY
Remote non-intrusive HR measurement is an interesting
topic for both commercial and academic purposes. Many
past works that attempt for remote pulse monitoring
include the utilization of photoplethysmography (PPG) [5,
9]. The blood volume of micro-vascular all over the body
changes together with cardiac pulse, so the blood volume
pulse (BVP) measured at peripheral body tissues (like
palm or fingertip) is usually used as an indicator of cycle
measurement. The principle of PPG method is to
illuminate the skin with a LED then measure the quantity
of light reflected or transmitted to a photodiode. Since the
quantity of light absorption may be a function of the blood
volume, PPG can measure the local blood volume pulse.
Although it's possible to use PPG based settings to live HR
with zero contact, this method still requires special
lighting sources and sensors. In the recent years a few
papers proposed shading based techniques for far off HR
estimation utilizing standard business cameras [10, 12,
13]. Poh et al. [12] investigated the likelihood to measure
HR from face recordings recorded by a web-cam.
Independent Component Analysis (ICA) was applied to
separate the PPG signal from the three colour traces, and
therefore the PPG signal was transferred into frequency
domain to seek out the frequency with the max power
greater then 0.7 Hz and less than 4 Hz as the HR
frequency. As indicated by past discoveries [20], the green
channel follow contains the most grounded
plethysmography signal among the three shading
channels. Poh's outcomes showed that contrasting with
the raw green follow, ICA isolated sources can accomplish
higher precision for estimating HR. The outcomes in [12]
were tested by Kwon et al. [10]. Kwon et al. recorded face
recordings with the inherent camera of an advanced cell,
and extricated HR utilizing both the crude green follow
and the ICA isolated sources. They observed that ICA
marginally dropped the presentation which is in
opposition to Poh's outcome. Later Poh et al. [13] worked
on their technique by adding a few transient channels both
when applying ICA. The better technique accomplished
exceptionally high exactness for estimating HR on their
self-gathered information. A movement-based technique
was proposed by Balakrishnan et al. as of late [2].
Balakrishnan et al. followed inconspicuous head motions
brought about via cardiovascular course, and utilized PCA
to separate the beat signal from the directions of different
followed highlight focuses. The strategy accomplished
promising execution on their self-gathered recordings.
Since the strategy depends on movement following,

2. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 03 | Mar 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 418
subjects should stay away from movement in their
analysis. Balakrishnan et al. shown that estimating HR on
moving subjects would be an important future bearing.
Sr
N
o.
Authors Name of the paper Description
1
A.
Asthana, S.
Zafeiriou,
S. Cheng,
and M.
Pantic
Robust
discriminative
response map
fitting with
constrained local
models
It presents a novel
discriminative
regression-based
approach for the
Constrained Local
Models (CLMs)
framework,
referred to as the
Discriminative
Response Map
Fitting (DRMF)
method, which
shows impressive
performance in
the generic face
fitting scenario.
2
G.
Balakrish
nan, F.
Durand,
and J.
Guttag
Detecting pulse
from head
motions in video
The results of this
work tracked
subtle head
oscillations caused
by cardiovascular
circulation, and
used PCA to
extract the pulse
signal from the
trajectories of
multiple tracked
feature points.
3
R. Basri
and D. W.
Jacobs.
Lambertian
reflectance and
linear subspaces.
It provides
algorithms for
recognition of
object with help of
linear methods
and algorithms
that enforce non-
negative lighting
functions. It also
shows a plain way
to impose
nonnegative
lighting when the
images of a body
lie near a 4D
linear space.
4
G. Bradski The OpenCV
Library.
It presents
information on a
study which
focused on
OpenCV, an open-
source, computer-
vision library for
drawing out and
processing
relevant data from
images.
5
G. Cennini,
J. Arguel,
K. Aks¸ it,
and A. van
Leest.
Heart rate
monitoring via
remote
photoplethysmogr
aphy with motion
artifacts
reduction.
The conclusion of
this paper was a
photoplethysmogr
aphic device that
operates remotely,
i.e., not in contact
with the skin.
6
K. Chan
and Y.
Zhang.
Adaptive
reduction of
motion artifact
from
photoplethysmogr
aphic recordings
using a variable
step-size lms filter.
Combination of
the ICA algorithm
and adaptive filter
to extract the
underlying PPG
signals from the
MA contaminated
PPG signals with
the amplitude
information
reserved.
7
F. X.
Gamelin,
S.
Berthoin,
and L.
Bosquet.
Validity of the
polar s810 heart
rate monitor to
measure rr
intervals at rest.
Comparing R-R
intervals and the
subsequent
analysis of heart
rate variability
(HRV) obtained
from the Polar
S810 heart rate
monitor (HRM).
8
M. H.
Hayes.
Recursive least
squares. Statistical
Digital Signal
Processing and
Modelling
It derived by
minimizing a
weighted least
squares error, and
derive an efficient
algorithm for
performing this
minimization
known as
recursive least
squares.
9
K.
Humphrey
s, T. Ward,
and C.
Markham.
Noncontact
simultaneous dual
wavelength
photoplethysmogr
aphy: a further
step toward
noncontact pulse
oximetry
The conclusion
was camera-based
gadget capable of
capturing two
photoplethysmogr
aphic (PPG)
signals at two
different
wavelengths at the
same time, in a
remote i.e., not in
contact with the
skin.
10 S. Kwon,
H. Kim,
and K. S.
Validation of heart
rate extraction
using video
It explores the
potential that the
authentic heart

3. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 03 | Mar 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 419
Park imaging on a built-
in camera system
rate can be
measured
remotely by the
facial video
recorded using
smartphone
camera.
11
C. Li, C. Xu,
C. Gui, and
M. D. Fox.
Distance
regularized level
set evolution and
its application to
image
segmentation.
The paper
registers a new
Different level set
composition in
which the
regularity of the
level set function
is inherently
maintained during
the level set
evolution.
12
M.-Z. Poh,
D. J.
McDuff,
and R. W.
Picard
Non-contact,
automated cardiac
pulse
measurements
using video
imaging and blind
source separation
It introduces a
new method for
remote
measurement of
pulse rate and can
provide reliable
assessment that
can be applied to
colour video
recordings of
human face.
13
M.-Z. Poh,
D. J.
McDuff,
and R. W.
Picard.
Advancements in
noncontact,
multiparameter
physiological
measurements
using a webcam.
They worked on
their technique by
adding a few
transient channels
both when
applying ICA. The
better technique
accomplished
exceptionally high
exactness for
estimating HR on
their self-gathered
information.
3. CONCLUSIONS
A real time noncontact based HR extraction method is
described in our paper using facial video which is easy to
implement, low cost easy to use for real time applications.
Here, the main idea is to extract HR from the colour
variation in the facial skin due to cardiac pulse and the
implementation has been done using a simple webcam in
indoor environment with constant ambient light. This non-
contact technology is promising for medical care and
others indoor applications due to widespread availability
of camera specially webcams. For applications in outdoor
environment for example driver monitoring, few things
such as variable environmental illumination or head
movement should be considered. Also, to increase the
efficiency, the experiment needs to be done by more test
subjects and more verifying systems.
Acknowledgement:
We wish to express our sincere gratitude to Dr. Shrikant
Kallurkar, Principal and Dr. Suvarna Pansambal, H.O.D.
of Department Computer Engineering of Atharva College
of Engineering for providing us an opportunity to do our
research work on “Pulse Rate Monitoring using Image
Processing”.
This research bears on imprint of many peoples. We
sincerely thank our project guide Prof. Jignesh Patel for his
guidance and encouragement in carrying out this synopsis
work.
Finally, we would like to thank our colleagues and friends
who helped us in completing project work successfully.
REFERENCES
[1] A. Asthana, S. Zafeiriou, S. Cheng, and M. Pantic. Robust
discriminative response map fitting with constrained local
models. In CVPR, 2013.
[2] G. Balakrishnan, F. Durand, and J. Guttag. Detecting
pulse from head motions in video. In CVPR, 2013.
[3] R. Basri and D. W. Jacobs. Lambertian reflectance and
linear subspaces. PAMI, 2003.
[4] G. Bradski. The OpenCV Library. Dr. Dobb’s Journal of
Software Tools, 2000.
[5] G. Cennini, J. Arguel, K. Aks¸it, and A. van Leest. Heart
rate monitoring via remote photoplethysmography with
motion artifacts reduction. Optics express, 2010.
[6] K. Chan and Y. Zhang. Adaptive reduction of motion
artifact from photoplethysmographic recordings using a
variable step-size lms filter. In Proceedings of IEEE on
Sensors, 2002.
[7] F. X. Gamelin, S. Berthoin, and L. Bosquet. Validity of
the polar s810 heart rate monitor to measure rr intervals
at rest. Medicine and Science in Sports and Exercise, 2006.
[8] M. H. Hayes. 9.4: Recursive least squares. Statistical
Digital Signal Processing and Modeling, 1996.
[9] K. Humphreys, T. Ward, and C. Markham. Noncontact
simultaneous dual wavelength photoplethysmography: a
further step toward noncontact pulse oximetry. Review of
scientific instruments, 2007.

4. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 03 | Mar 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 420
[10] S. Kwon, H. Kim, and K. S. Park. Validation of heart
rate extraction using video imaging on a built-in camera
system of a smartphone. In EMBS, 2012.
[11] C. Li, C. Xu, C. Gui, and M. D. Fox. Distance regularized
level set evolution and its application to image
segmentation. IEEE Trans. on Image Processing, 2010.
[12] M.-Z. Poh, D. J. McDuff, and R. W. Picard. Non-contact,
automated cardiac pulse measurements using video
imaging and blind source separation. Optics Express,
2010.
[13] M.-Z. Poh, D. J. McDuff, and R. W. Picard.
Advancements in noncontact, multiparameter
physiological measurements using a webcam. IEEE Trans.
on Biomedical Engineering, 2011.