Items with no label
3335 Discussions

Do D435/415 and Sdk 2.x support heart rate monitoring?

AAhme16
Beginner
1,969 Views

Hi,

I am planning to buy one of the newest Intel realsense cameras (D435/415) for a health monitoring project.

But, I couldn't find a similar documentation for SDK 1.x for pulse estimation like in this link (for F200/SR300 cameras).

https://software.intel.com/sites/landingpage/realsense/camera-sdk/v1.1/documentation/html/index.html?doc_face_pulse_estimation_data.html Intel® RealSense™ SDK 2016 R2 Documentation

Do D435/415 and SDK 2.x support heart rate monitoring?

Thanks!

0 Kudos
1 Solution
MartyG
Honored Contributor III
548 Views

The 'Pulse Estimation' feature, and similar functions that developers have created, work by measuring changes in the color of the skin, since the blood flows under the skin. There is not a pre-made estimation function in SDK 2.0 to the best of my knowledge. You may be able to integrate a pre-made pulse calculation module into an SDK 2.0 application though, since the SDK supports integration with a range of programming languages and software platforms.

For example, here is a webcam pulse detector module in the Python language, which SDK 2.0 supports.

https://github.com/thearn/webcam-pulse-detector GitHub - thearn/webcam-pulse-detector: A python application that detects and highlights the heart-rate of an individual …

OpenCV pulse detectors tend to use analysis based on face detection via video frames.

Example 1:

https://github.com/shelhamer/visual-pulse-monitor GitHub - shelhamer/visual-pulse-monitor: Measure heart rate from standard video recording (UMass CS 691A final project)

Example 2:

http://jhnet.co.uk/projects/heart_monitor Webcam Heart-Rate Monitor - Projects - Jhnet

You can research this subject further by googling for 'calculate pulse skin color'.

View solution in original post

4 Replies
MartyG
Honored Contributor III
549 Views

The 'Pulse Estimation' feature, and similar functions that developers have created, work by measuring changes in the color of the skin, since the blood flows under the skin. There is not a pre-made estimation function in SDK 2.0 to the best of my knowledge. You may be able to integrate a pre-made pulse calculation module into an SDK 2.0 application though, since the SDK supports integration with a range of programming languages and software platforms.

For example, here is a webcam pulse detector module in the Python language, which SDK 2.0 supports.

https://github.com/thearn/webcam-pulse-detector GitHub - thearn/webcam-pulse-detector: A python application that detects and highlights the heart-rate of an individual …

OpenCV pulse detectors tend to use analysis based on face detection via video frames.

Example 1:

https://github.com/shelhamer/visual-pulse-monitor GitHub - shelhamer/visual-pulse-monitor: Measure heart rate from standard video recording (UMass CS 691A final project)

Example 2:

http://jhnet.co.uk/projects/heart_monitor Webcam Heart-Rate Monitor - Projects - Jhnet

You can research this subject further by googling for 'calculate pulse skin color'.

AAhme16
Beginner
548 Views

Thanks, MartyG.

I have seen the examples that you posted and both project work on a simple webcam. not the intel realSense Camera.

If I use the D400 series camera, will I get additional accuracy? what are exactly the benefits of using realSense Camera in such project?

Thanks.

0 Kudos
MartyG
Honored Contributor III
548 Views

The 400 Series cameras have significant technical advantages over an ordinary webcam. For example, if your project involves depth data, the 400 Series has a depth sensing range of 10 m thanks to its D4 Vision Processor component. The cameras can also run up to 90 frames per second in depth mode. Furthermore, the D4 Vision Processor enables an error rate of less than 1%.

0 Kudos
AAhme16
Beginner
549 Views
0 Kudos
Reply