Pyroelectric IR sensor arrays for fall detection in the older population

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.

Fall Detection Application on an ARM and FPGA Heterogeneous Computing Platform

“… Indoor sensors: Arrays of infrared sensors [1], laser scanners [2] and even floor vibration and sound [3].  Wearable sensors: Accelerometers, gyroscopes or even barometric pressure sensors [4].…”

“…Background subtraction method is applied in this study. A pixel is marked as foreground If |I i -B i | > τ, (1) Where I i is current video frame; B i is modelled background frame and τ is a "predefined" threshold. The updated background is estimated as:…”

Radio tomographic imaging based body pose sensing for fall detection

“…Besides, camera-based methods expose the privacy of the older adults in free-living environments. A number of alternative solutions have arisen to analyze human motion with isomorphism thermal cameras (Han and Bhanu 2005;Sixsmith et al 2005). The human body is considered as a natural emitter of infrared ray.…”

Person Authentication and Activities Analysis in an Office Environment Using a Sensor Network

“…Person surveillance and activity analysis have been applied to various circumstances, such as security control, person tracking, assisted living and humancomputer interface (HCI), etc [1][2][3][4][5][6]. In recent years, along with the rapid development of computer vision and many kinds of network devices, person authentication, tracking and human-behavior understanding have become indispensable for providing many kinds of personalized services in response to the implicit/explicit demands of the users.…”