CARDIOVASCULAR SIGNAL RATE DETECTION USING FREQUENCY DOMAIN TRANSFORM

Abstract
Cardiovascular Rate (HR) is a crucial indicator of people’s physiological state therefore it is important to monitor often. It is an essential parameter in most of the medical diagnosis for innumerable medical condition. This paper deals with the diagnosis of heart problems by computing the heart rate measurements from human facial videos recorded using a simple webcam. The main target is to extract cardiovascular rate from the skin color variation in the facial tissues caused due to oxygen supply all over the body. The face detection is carried out using Viola Jones algorithm which scans a subwindow capable of detecting faces across a given input image based on Haar-like rectangular features that are extracted from integral images. The forehead region is selected for Green channel estimation then Empirical mode decomposition is performed for reflectance decomposition of the forehead region. Finally heart rate is estimated by detecting power spectral density of frequency domain images. Obtained PSD is compared with standard heart rate to obtain the subject’s cardiovascular rate.

Authors
J Hareen Breath Mary, C Rekha
PET Engineering College, India

Keywords
Viola Jones Algorithm, Haar-like Rectangular Feature, Reflectance Decomposition, Green Channel, PSD
Published By :
ICTACT
Published In :
ICTACT Journal on Image and Video Processing
( Volume: 9 , Issue: 4 )
Date of Publication :
May 2019

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.