An acoustic fall detection system using beamforming, source separation and Kinect
Metadata[+] Show full item record
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] In this dissertation, we first review the previous work on acoustic fall detection system based on the contributed paper work. Then we propose a Kinect-based fall detection system which significantly improves the performance using beamforming and source separation techniques. The latest acoustic fall detection system (acoustic FADE) has achieved encouraging results on real-world dataset. However, the acoustic FADE device is difficult to be deployed in real environment due to its large size. In addition, the estimation accuracy of sound source localization (SSL) and direction of arrival (DOA) becomes much lower in multi-interference environment, which will potentially result in the distortion of the source signal using beamforming (BF). Microsoft Kinect is used to address these issues by measuring source position using the depth sensor. We employ robust minimum variance distortionless response (MVDR) adaptive BF (ABF) to take advantage of well-estimated source position for acoustic FADE. A significant reduction of false alarms and improvement of detection rate are both achieved using the proposed fusion strategy on real-world data. In addition, we propose an unsupervised-learning-based single-channel source separation approach for removing the interference. Then we propose a multi-channel source separation by averaging the results obtained by the proposed single-channel approach. The prototype of Kinect fall detection using source separation is also proposed. The experiments show that the best fall detection performance is achieved by using the proposed multi-channel source separation.
Access is limited to the campuses of the University of Missouri.