Japanese firm Weathernews has developed an artificial intelligence (AI) based highway management support system for real-time recognition of road surface snow.

Known as the AI Road Management Support System, the technology is said to be the world’s first such solution.

After analysing photos and video obtained from fixed-point cameras, the system automatically identifies road surface conditions in real-time.

In this system, the AI has been educated using machine learning to perceive differences in image data between accumulated and packed snow, slush and dry roads.

As a result, the system is capable of recognising intricate road conditions where different types of snow are mixed together.

“We will push forward with development, aiming to have it ready for implementation this summer, so that it can be put into use next winter season.”

Weathernews claims that the system can precisely detect road surface conditions even if a car is driven through a snowstorm at night, which is very difficult with conventional image analysis methods.

The company claimed that the new system will allow for increased efficiency for snow and ice countermeasure works, as well as helping to cut traffic collisions and improve self-driving technology.

The company said in a statement: “Initially, we will push forward with development, aiming to have it ready for implementation this summer, so that it can be put into use next winter season by road managers and municipalities for their work on snow and ice countermeasures.”

In the future, the company is planning to combine not just video and still photos from fixed cameras, but also media and comments on road surface conditions sent to it by weather reporters.