Controlling urban infrastructure, including tracking the state of traffic signs and identify the need of repair plates, is a rather laborious process. At the moment this could be the working staff, but thanks to Google Street View and artificial intelligence, this process can be automated. The technology is already established and has proved its ability to accurately identify any damaged road signs, but there is one problem that almost driving the use of technology to nothing.
The essence of the technology is that the trained artificial intelligence validates the photos, made especially for Google Street View, for damaged road signs. These images are captured a specially equipped car since the day of establishment of the service in 2007 — they drive around on roads around the world and shoot them on camera so people could watch on the streets of cities and find the right place simply by logging online.
Artificial intelligence recognizes a broken traffic signs
Artificial intelligence was developed by the staff of RMIT University, and has demonstrated very good results. Studying an extensive database of Street View images, he was able to recognize traffic signs with 96% accuracy, and to determine the need for repair up to 98%. Thanks to artificial intelligence, and geo-data provided by broken signs, workers can quickly learn about them and promptly repaired.
The main disadvantage of Google Street View
The technology seems ideal, and in theory can reduce the number of road accidents. However, there is one problem — the images in Google Street View are updated very rarely, because cars rarely pass the same place several times with different intervals. Of course, the company may establish a weekly, or even daily plying on the same routes, but it will be worth a lot of money.
At the moment there is one potential solution to this problem. The developers of the monitoring system of urban infrastructure can equip the cameras of the Street View cars that travel daily on the same route. For example, a tool for regular updates of the database can become garbage, because every morning, run on the same routes.
The service Street View allows you to visit the most unusual places in the world, without even leaving home. For example, users can look at Devon island, the largest uninhabited island in the world, and its landscape is most similar to the Martian. To read more about this unusual place and view his panoramic images can be in our material.
If you are interested in news of science and technologies, be sure to subscribe to our channel in Yandex.Zen. There you will find materials that were not published on the website!