Numéro |
J. Phys. IV France
Volume 128, September 2005
|
|
---|---|---|
Page(s) | 153 - 160 | |
DOI | https://doi.org/10.1051/jp4:2005128024 |
J. Phys. IV France 128 (2005) 153-160
DOI: 10.1051/jp4:2005128024
Pyroelectric IR sensor arrays for fall detection in the older population
A. Sixsmith1, N. Johnson2 and R. Whatmore31 Department of Primary Care, Whelan Building, Brownlow Hall, University of Liverpool, Liverpool L69 3GB, UK
2 IRISYS Ltd., The Mill, Towcester, UK
3 Department of Advanced Materials, Cranfield University, Cranfield, Beds MK43 0AL, UK
Abstract
Uncooled pyroelectric sensor arrays have been studied over many
years for their uses in thermal imaging applications. These arrays will only
detect changes in IR flux and so systems based upon them are very good at
detecting movements of people in the scene without sensing the background, if
they are used in staring mode. Relatively-low element count arrays (16 x 16) can
be used for a variety of people sensing applications, including people counting
(for safety applications), queue monitoring etc. With appropriate signal
processing such systems can be also be used for the detection of particular
events such as a person falling over. There is a considerable need for automatic
fall detection amongst older people, but there are important limitations to some
of the current and emerging technologies available for this. Simple sensors,
such as 1 or 2 element pyroelectric infra-red sensors provide crude data that is
difficult to interpret; the use of devices worn on the person, such as wrist
communicator and motion detectors have potential, but are reliant on the person
being able and willing to wear the device; video cameras may be seen as
intrusive and require considerable human resources to monitor activity while
machine-interpretation of camera images is complex and may be difficult in this
application area. The use of a pyroelectric thermal array sensor was seen to
have a number of potential benefits. The sensor is wall-mounted and does not
require the user to wear a device. It enables detailed analysis of a subject's
motion to be achieved locally, within the detector, using only a modest
processor. This is possible due to the relative ease with which data from the
sensor can be interpreted relative to the data generated by alternative sensors
such as video devices. In addition to the cost-effectiveness of this solution,
it was felt that the lack of detail in the low-level data, together with the
elimination of the need to transmit data outside the detector, would help to
avert feelings intrusiveness on the part of the end-user. The main benefits of
this type of technology would be for older people who spend time alone in
unsupervised environments. This would include people living alone in ordinary
housing or in sheltered accommodation (apartment complexes for older people with
local warden) and non-communal areas in residential/nursing home environments
(e.g. bedrooms and ensuite bathrooms and toilets). This paper will review the
development of the array, the pyroelectric ceramic material upon which it is
based and the system capabilities. It will present results from the Framework 5
SIMBAD project, which used the system to monitor the movements of elderly people
over a considerable period of time.
© EDP Sciences 2005