Filtern
Erscheinungsjahr
- 2017 (2) (entfernen)
Sprache
- Englisch (2) (entfernen)
Schlagworte
- Digitale Bildverarbeitung (2) (entfernen)
Institut
- Sonstige (2) (entfernen)
Road condition acquisition and assessment are the key to guarantee their permanent availability. In order to maintain a country's whole road network, millions of high-resolution images have to be analyzed annually. Currently, this requires cost and time excessive manual labor. We aim to automate this process to a high degree by applying deep neural networks. Such networks need a lot of data to be trained successfully, which are not publicly available at the moment. In this paper, we present the GAPs dataset, which is the first freely available pavement distress dataset of a size, large enough to train high-performing deep neural networks. It provides high quality images, recorded by a standardized process fulfilling German federal regulations, and detailed distress annotations. For the first time, this enables a fair comparison of research in this field. Furthermore, we present a first evaluation of the state of the art in pavement distress detection and an analysis of the effectiveness of state of the art regularization techniques on this dataset.
Cyclists are more likely to be injured in fatal crashes than motorised vehicles. To gain detailed and precise behavioural data of road users, i.e. trajectories, a measuring campaign was conducted. Therefore, a black-spot for accidents with cyclists in Berlin, Germany was selected. The traffic has been detected by a fully automated traffic video analysis system continuously for twelve hours. The video surveillance system is capable of automatically extracting trajectories, classifying road user types and precise determining and positioning of conflicts and accidents. Additionally, pre-conflict and pre-accident situations could be analysed to provide further in-depth understanding of accident causation. The evaluation of the measuring campaign comprised the investigation of traffic parameters, e.g. traffic flow, as well as traffic-safety related parameters based on Surrogate Safety Measures (SSM). Furthermore, the spatial and temporal distributions of conflicts involving cyclists were determined. As a result, three possible conflict clusters could be identified, of which one cluster could be confirmed by detailed video analysis, showing conflicts caused by right turning vehicles.