Video can provide an important source of information for Driver Assistance Systems. Information about the scene ahead of the vehicle, such as the location of lane markings and obstacles can lead to significant improvements in system performance. However, algorithms for processing video data can be extremely complex and may have large computational requirements if they are to be executed in real-time. In this paper we present an embedded image processing platform which allows the design, development and partitioning of algorithms prior to production-intent design. The platform makes use of a powerful DSP and an FPGA that provides significant computational ability while remaining easily embeddable in a vehicle. The system also contains a CAN controller, inputs for up to 3 Camera Link cameras and an SVGA output. The system is embedded, modular and extensible and is therefore fit for both bench-top and in-vehicle algorithm development and demonstration.
An embedded image processing platform for driver assistance system algorithm development
Eine gemeinsame Bildverarbeitungs-/Programmentwicklungsplattform für Fahrerassistenzsysteme
2001
8 Seiten, 4 Bilder, 5 Quellen
Aufsatz (Konferenz)
Englisch
An embedded image processing platform for driver assistance system algorithm development
British Library Conference Proceedings | 2001
|Image processing for driver assistance
Tema Archiv | 1998
|An image processing system for driver assistance
British Library Online Contents | 2000
|Driver-assistance device, driver-assistance system, and driver-assistance program
Europäisches Patentamt | 2022
|