Contactless Heart Rate Monitor for Multiple Persons in a Video
In this project, we have proposed an algorithm to develop a contactless Heart Rate Monitor using video for multiple people. The video is acquired at a rate of 30 fps with the resolution of 640x480. Firstly, the Viola Jones face detector will be used for face detection purposed. After that, the regio...
| Main Author: | |
|---|---|
| Format: | Final Year Project / Dissertation / Thesis |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.utar.edu.my/1795/ http://eprints.utar.edu.my/1795/1/Contactless_Heart_Rate_Monitor_for_Multiple_Persons_in_a_Video.pdf |
| Summary: | In this project, we have proposed an algorithm to develop a contactless Heart Rate Monitor using video for multiple people. The video is acquired at a rate of 30 fps with the resolution of 640x480. Firstly, the Viola Jones face detector will be used for face detection purposed. After that, the regions of interest (cheeks) are detected with the coordinate geometry algorithm. Then 3 color traces are obtained for every image by applying mean on every single color plane. Fast Fourier Transform will be applied to these traces and peak frequency will be detected after band pass filtering. The conversion between frequencies to bpm will be done. It has been seen that green channel of RGB color modal given the best result for heart rate. The accuracy of heart rate detection for single person is 7.5% and the multiple persons is 11.3%. |
|---|