I am currently researching the use of Information and Communication Technologies as tools for preventative care and diagnosis support in elderly population. Our current approach uses accelerometers and video sensors for the recognition of instrumental daily living activities (IADL, e.g., preparing coffee, making a phone call). Clinical studies have pointed the decline of elderly performance in IADL as a potential indicator of early symptoms of a dementia case (e.g., Alzheimer’s’ patients). IADL are modeled and detected using a constraint-based generic ontology (called SCREK). This ontology allows us to describe events based on spatial, temporal, and sensors data (e.g., MotionPOD) values.
The event recognition is based on the human pre-definition of scenarios of interest corresponding to the events to be recognized. Scenarios are written in a formal language easily readable by humans. An event (or scenario) may involve detected objects (people, vehicle, groups…) and contextual objects (walls, equipment…) or zones. Detected objects arrive as a metadata stream from other algorithms (people tracking, group tracking…).
Issues of event recognition mainly concern the uncertainty of the input objects detection.