Now, however, engineering scholars at the Canada’s University of Waterloo have built Application that can tell if the Driver is meshing in distracting actions such as texting.
Developed by a group led by Prof. Fakhri Karray at the university’s Centre for Pattern Analysis and Machine Intelligence (CPAMI), the method uses in-car cameras and computer algorithms to watch for hand actions that vary from those commonly linked with driving. Those algorithms, which were “taught” using machine-learning methods, are able to coordinate the actions to known actions like texting, having a cell phone communication, or stretching over into the back seat.
Once recognized, distracting actions are ranked according to the protection threat that they pose this is based not just on what the activity is, but also on parts such as its term. In cases where the disturbance level is considered to be too high, the method could warn the driver, or in severe cases even momentarily take over the driving of the car.
Karray and colleagues are now studying at combining other distracted-driving signs into the system, including the monitoring of top and face position.
They recently did their research at the 14th International Conference on Image Analysis and Recognition in Montreal.