APPLICATION FOR THE AUTOMOTIVE SECTOR
How can each one of us help to optimise the performance of driving assistance systems in order to consistently reduce the accident rate?
While driverless vehicles still have a long way to go before being a standard on the road, integrated systems which support the driver are already part of everyday life. From the lane-keeping assistant which independently intervenes in extreme danger situations, as in the case of a second sleep, to systems which assume routine work, such as parking. The driver is still fully responsible and high requirements must be met prior to approval of such systems.
The Berlin-based company Hella Aglaia is the market leader in the testing, development, and certification of safety-relevant driver assistance systems. In order to test the performance of driving assistance systems and to train their algorithms, the human being is still the top corrective. It defines the action rule that the algorithm performs. The example of driver assistance systems, therefore, requires a lot of video material to be screened and prepared. With the aid of a software, relevant objects in road traffic are marked and commented image sequence by image sequence. These are referred to as so-called “ground truth” reference data and are used in order to be able to match them with the results of the image processing algorithms. This process is called “labeling” and is performed by so-called “labelers”.
In order to create optimised and user-friendly workflows as well as to sustainably reduce the currently high label costs, UseTree supports Hella Aglaia in various projects as a UX partner – for more safety in road traffic.
HELLA Aglaia Mobile Vision