Sonstige
Filtern
Sprache
- Englisch (2) (entfernen)
Volltext vorhanden
- nein (2) (entfernen)
Schlagworte
- Detektion (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.
It is well known that most accidents with pedestrians are caused by the driver not being alert or misinterpreting the situation. For that reason advanced forward looking safety systems have a high potential to improve safety for this group of vulnerable road users. Active pedestrian protection systems combine reduction of impact speed by driver warning and/or autonomous braking with deployment of protective devices shortly before the imminent impact. According to the Euro NCAP roadmap the Autonomous Emergency Braking system tests for Pedestrians Protection will be set in force from 2016 onwards. Various projects and organisations in Europe are developing performance tests and assessment procedures as accompanying measures to the Euro NCAP initiative. To provide synthesised input to Euro NCAP so-called Harmonisation Platforms (HP-) have been established. Their main goal is to foster exchange of information on key subjects, thereby generating a clear overview of similarities and differences on the approaches chosen and, on that basis, recommend on future test procedures. In this paper activities of the Harmonisation Platform 2 on the development of Test Equipment are presented. For the testing targets that mimic humans different sensing technologies are required. A first set of specifications for pedestrian targets and the propulsion systems as collected by Harmonisation Platform 2 are presented together with a first evaluation for a number of available tools.