University of Houston
Department of Computer Science
In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy

Jin Fei

will present his dissertation

Breathing Computation through Thermal Imaging

Abstract

This dissertation studies human breathing rate and waveform through thermal imaging, which could indicate the overall health status of a person. Based on the facial radiation information, we track the face, localize the nasal region, and extract the breathing signal. It is quasi-periodic due to the interleaving of high and low temperatures corresponding to expiration and inspiration, respectively. Originally, we compute the signal frequency and analyze the waveform by Fast Fourier Transform. Recently, we apply Wavelet Transform to analyze both breathing descriptors.

We have performed experiments on both laboratory and clinical subjects. The computed breathing rate from thermal imaging is juxtaposed with traditional contact devices: abdominal transducer belt and thermistor. The harmonic components could characterize breathing waveform. Our thermal breathing signals are highly correlated with the temperature signals captured by thermistor. In the clinical studies, we can detect sleep apnea. The contact-free and passive thermal imagery opens the way for clinical practice. It could find widespread applications in preventive medicine as well as sustained physiological monitoring of subjects.

Date: Thursday, November 8, 2007
Time: 11:30 AM
Place: 218-PGH

Faculty, students, and the general public are invited.
Dissertation Advisor: Dr. Ioannis Pavlidis