Address-Event Fall Detector for Assisted Living Applications

In collaboration with Tobi Delbruck, Institute for Neuroinformatics (INI), Switzerland.

In this project we profile a fall detector using an asynchronous temporal contrast vision sensor. Figure illustrated the fall detector setting in assisted living applications. The detectors take the side-view of the monitored person. It can detect accidental activities and raise alarms in elderly home care environments. The vision systems are mounted on the wall at a height of 0.8m, which is approximately the same height of a light switch.

 

 

 

Address-event fall detectors are used for assisted living applications. The detectors are mounted on the wall at a height of 0.8m, which is approximately the same height of a light switch.




Why Fall Detection?

Fall is a major health hazard for the elders when they live independently. Approximately 30% of 65-year-old people fall each year. This number becomes higher in medical service institutions. Although less than one fall in 10 results in an injury, a fifth of fall incidents require medical attention. Another recent publication indicates that 50% of patients in nursery institutions fall each year, while 40% of them fall more than once. How to effectively assess, respond and assist elderly patients in trouble becomes an important research topic in medical elderly care services.

The 64x64 address-event (AE) temporal-constrast vision sensor used in the fall detector system

(a) Temporal contrast image from the ATC image sensor and (b) its targeted scene. The imaging system is placed in front of the subject with a distance of 3 meters and a height of 0.8 meter.
Here is the videos of the fall detector.

Centroid address-event evaluation matrix of five common home-assisted living scenarios. The centroid vertical velocities are normalized for distance.

AER Emu: an frame-based Address-Event image sensor emulator. This software project is in collaborated with Yale ENALAB



Download

People