Paving-map as a prototype based on AI

Abstract

As a part of The Danish Agency for Data Supply and Efficiency (SDFE)'s focus and goal of using artificial intelligence and creating a coherent data base, a prototype is presented on a high-resolution overview of paving-areas.

The map of the paving areas was created using the machine learning algorithm RandomForests on GeoDanmark's spring orthophoto. The plan is to include other data sources such as LiDAR or summer orthophotos as well as various types of filters to strengthen the classification of algorithm.

The prototype will be presented at SDFE LABS, which is a new platform for introducing and user dialogue about candidates for new and free data sets. Feedback is desired to further develop the prototype in a direction that matches the users' need.