State-of-the-art advanced driver assistance systems (ADAS) typically focus on single tasks and therefore, have functionalities with clearly defined application areas. Although said ADAS functions (e.g. lane departure warning) show good performance, they lack general usability, as e.g. different modes of operation for highways and country roads. This paper presents a real-time capable approach, which classifies the driving scene by using the newly developed Hierarchical Principal Component Classification (HPCC). Based on that, an ADAS gets information about the current scene context and is able to activate different operation modes. Exemplarily, the algorithm was trained on three different categories (highways, country roads, and inner city), but can be applied to any number and type of categories. Evaluation results on 9000 images show the reliability of the approach and mark it as a crucial step towards more sophisticated high level applications.
Image-based classification of driving scenes by Hierarchical Principal Component Classification (HPCC)
2009 IEEE Intelligent Vehicles Symposium ; 341-346
01.06.2009
5450175 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
British Library Conference Proceedings | 2009
|Principal component analysis-based learning for preceding vehicle classification
IET | 2014
|Principal component analysis‐based learning for preceding vehicle classification
Wiley | 2014
|