Mobile Cardiac Pulse Measurements

Cardiac pulse measurements play an important role in diagnosing heart diseases. In this paper, we describe a methodology that simply depends on a series of facial images. By performing automatic face tracking and blind source separation of three color channels into independent components, we extract the blood volume pulse from the recorded facial frames. Heart rates can be subsequently quantified from previous results. The introduced method is featured as low-cost, automatic, absolutely contact-free, and motion-tolerant. We implement the measuring platform on a PC and web first, and then realize an Android phone version using server-client communication.