摘要：Heart rate is an important indicator for the mental and physical state, but it is usually measured through physical contact. In this paper, a novel non-contact method of heart rate measurement has been proposed from the infrared sequence images. First, a square region of interest (ROI) was manually selected to cover the temple on the first frame. Then a head movement detection algorithm based on the centroid coordinate change of the extracted skin area was applied to the infrared sequence images, and according to the movement result, the ROI location in subsequent frames could be identified. After that, the distance between the gravity center and the top-left corner in every ROI is computed to get a time-lapse signal. Finally, the discrete wavelet transform and an autoregressive model were used respectively to recognize the heart rate. Fourteen healthy subjects (24-29 years of age, 4 females and 10 males) participated in the experiment. Compared with the concomitant ECG, mean accuracy rate of 94.5% was acquired. The results show the potential of our method for non-contact heart rate measurement.
关键词：heart rate;infrared sequence images;wavelet transform;autoregressive model
