Video-based driver's assistant systems will play a major role in future car safety system Previous work of the Daimier-Benz Research in the Prometheus project has shown that functions such as lane keeping, object detection and avoidance, and traffic sign recognition can be performed in real-time. However, conventional image processing architectures consisting of CCDs and powerful digital computers are too heavy, too bulky, consume too much power, and are too expensive. Moreover, because of some considerable deficits of CCDs, the lack of processing speed and robustness, these architectures cannot be transfered into products. Image processing chips and architectures in highly parallel analogue CMC technique will be proposed. With these systems, it will be possible to realize miniaturized, cheep, and powerful driver's assistant systems. This work has shown that highly parallel, powerful, robust, and cheap vision systems for driver's assistant systems can be built on the basis of analogue vision chips. Several types of algorithms are developed which could not yet be used in digital image processing because of the computational load. Implemented in a chip, they solve the perceptual task by a non-linear system of recursive equations. Typical relexation times are in the order of microseconds; however, in the current architecture the data transfer times between the chips limit the frame rate to 50 Hz. The algorithm development has to meet the requirements from applied image processing as well as the constraints of chip manufacturing and system integration. In order to produce chips with many nodes, modern sub-micron-technology must be used. The advantage of analogue vision chips will only then be fully exploited, when the most modern technologies for packaging, integration, and 3d-connection will be used.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Analogue image processing for driver's assistant systems


    Additional title:

    Analoge Bildauswertung für Fahrer-Unterstützungstechnologien


    Contributors:
    Zinner, H. (author) / Nothaft, P. (author)


    Publication date :

    1996


    Size :

    14 Seiten, 5 Bilder


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    AIR BAG DEVICE FOR ASSISTANT DRIVER'S SEAT

    YAMADA IKUO | European Patent Office | 2017

    Free access

    Realization of a Driver's Warning Assistant for Intersections

    Enkelmann, W. / Nirschl, G. / Gengenbach, V. et al. | British Library Conference Proceedings | 1993


    Realization of a driver's warning assistant for intersections

    Enkelmann, W. / Nirschl, G. / Krüger, W. et al. | Tema Archive | 1993



    Real-time vision-based intersection detection for a driver's warning assistant

    Rössle, S. / Krüger, W. / Gengenbach, V. | Tema Archive | 1993