Neuroinformatics The driving school for computers
In order to generate realistic images of road signs, researchers pit two algorithms against each other.
To ensure that cars will one day drive autonomously and safely through the streets, they must be able to recognise road signs. Even at night, in the rain, in the snow, or if the signs are covered in moss, dirty or partially overgrown. In order to learn how to do that, they require a plethora of examples of all road signs from different seasons, times of day and weather conditions. Together with Dominic Spata und Daniela Horn, Professor Sebastian Houben from the RUB Neural Computation Institute has therefore developed a method to generate traffic signs automatically that computers can use to practise vision.
“We want to reach a point where an algorithm learns to generate images of road signs that other programs can use to practise their recognition capabilities,” elaborates Sebastian Houben. The research team uses two algorithms for this purpose: one is fed pictograms of official road signs and is given the task to generate images that look like photos; plus, the algorithm must be able to recognise the original sign in those images at a later point. “This is how we prevent the algorithm from distorting the image of the sign to such an extent that it no longer resembles the road sign in any way,” explains Daniela Horn.
Algorithms are sparring partners
The second algorithm has to decide if the generated image is a real photo or not. The goal is to ensure that the second algorithm can no longer tell what it is. “Moreover, the second algorithm indicates to the first one in what way the selection process could be made even more difficult,” says Sebastian Houben. “Accordingly, these two are sparring partners, of sorts.”
Checking in after a few days
At first, the training process doesn’t work particularly well. It counts as a success if the picture of a priority road sign has the right colour and is more or less square. But it’s improving apace. “After two or three days, we check what the pictures of road signs look like,” explains Daniela Horn. “If the pictures don’t look good to our human eye, we modify the algorithm.”