dc.contributor.advisor | Tyrer, Harry W. | eng |
dc.contributor.author | Muheidat, Fadi | eng |
dc.date.issued | 2015 | eng |
dc.date.submitted | 2015 Fall | eng |
dc.description.abstract | According to the World Health Organization-WHO, a fall is defined, as "inadvertently coming to rest on the ground, floor or other lower level, excluding intentional change in position to rest in furniture, wall or other objects". It is known that a senior who falls is at risk for serious injury and after necessary spiraling down eventually to death. Researchers concern is to develop new technology or enhance existing one to detect falls and reduce the consequences of a fall. We enhanced smart carpet, which is a floor based personnel detector system, to detect falls using a faster but low cost processor. Our new hardware front end reads from 128 sensors (the sensors output a voltage due to a person walking or falling on the carpet). The processor is Jetson TK1, which provides more computing power than before. We generated a dataset with volunteers who walked and fell to test our algorithms. Data Obtained allowed examining data frames read from the data acquisition system. We used different algorithms and techniques, and varied the windows size of number of frames (WS>=1) and threshold (TH). We found that at (WS=8), and threshold (TH=8) using connected component labeling algorithm (CCL) produced a fall sensitivity of 87.9%. We then used the dataset obtained from applying a set of fall detection algorithms and the video recorded for the fall patterns experiments to train a set of classifiers using multiple test options using the Weka framework. We found that the widow size (WS=8) at a threshold (TH=8) using connected component algorithm generated attribute contributed to the fall sensitivity. We measured the performance of each testing options. The best feature was again the size of the connected component with WS=8, with classification accuracy of 96.94%. Other algorithms attributes did not contribute significantly to the detection of the fall. | eng |
dc.identifier.uri | https://hdl.handle.net/10355/60033 | |
dc.language | English | eng |
dc.publisher | University of Missouri--Columbia | eng |
dc.relation.ispartofcommunity | University of Missouri--Columbia. Graduate School. Theses and Dissertations | eng |
dc.source | Submitted by University of Missouri--Columbia Graduate School. | eng |
dc.subject | fall detection, reducing fall consequences, smart carpet | eng |
dc.title | Adding intelligence to a floor based array personnel detector | eng |
dc.type | Thesis | eng |
thesis.degree.discipline | Computer engineering (MU) | eng |
thesis.degree.grantor | University of Missouri--Columbia | eng |
thesis.degree.level | Masters | eng |
thesis.degree.name | M.S. | eng |