Inventory of road characteristics

Rijkswaterstaat

Customer

Rijkswaterstaat

TECHNOLOGY

Sentinizer™

Sentinizer™ is Bolesian’s platform for inspecting visual data and automatically detect, classify and locate anomalies or objects. Sentinizer is often used to optimize infrastructure asset management. One of the applications of Sentinizer is road surface markings. Sentinizer can detect these markings, classify the type, analyze the current state (pollution and damage or wear) and report findings and geo-data. In this specific case for Rijkswaterstaat we used Sentinizer in a slightly different way …

Rijkswaterstaat is the government organization responsible for the design, construction, management and maintenance of the main infrastructure facilities in the Netherlands. One of the services of Rijkswaterstaat is the National Road Database (NWB – Nationaal Wegen Bestand). This database contains data of all roads, including numerous features such as traffic lanes, signage and other road-related objects.

Until now, updating the road file has always been done manually based on video recordings. Inspectors view the recordings and note all relevant road features they see. You can imagine that this is a very labor-intensive job to perform frequently and keep the database up to date. If the execution of a task requires human intelligence, but the handling of the task is inhumanly large, the deployment of artificial intelligence is valuable. 

Commissioned by Rijkswaterstaat, we deployed Sentinizer to process realtime video feeds and make an inventory of all kinds of traffic lanes based on road markings. Sentinizer detects road surface markings on the recordings and classifies the type of marking. After detection and classification knowledge rules are applied to determine the number and types of traffic lanes.

In the video you can see (maximize the window with the button on the right) how the markings are detected. The findings are annotated at the bottom of the screen. Here you will see the detected traffic lanes (rijstroken), the insertion and exit lanes (mengstroken), the markings and the types of markings on the left and the right.