Refine
Document type
- Article (peer-reviewed) (3)
- Conference Proceeding (2)
- Contribution to a Periodical (2)
- Report (2)
- Other (1)
Has full text
- No (10)
Keywords
- Automatische Erfassung (1)
- Bicycle paths (1)
- Bilderkennung (1)
- Fahrradsicherheit (1)
- Fahrradwege (1)
- Image recognition (1)
- Machine learning (1)
- Oberflächenbeschaffenheit (1)
- Online-Ressource (1)
- Smartphone-generated video data (1)
In this paper, a multi-stage approach is presented that uses machine learning methods to determine the surface condition of bicycle lanes. This approach uses a low-threshold acquisition method through smartphone-generated video data for image recognition to automatically detect and classify the surface condition and its state. The results are displayed on digital maps to increase their level of detail. The aim is to improve the safety and comfort of cyclists and to enable better maintenance of the paths.
Wie gut sind unsere Radwege?
(2022)