Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Automatic flow control planning for real-time image processing devices

: Melchert, W.


IEEE Signal Processing Society:
International Conference on Image Processing 1996. Proceedings. Vol.3 : September 16 - 19, 1996, Lausanne, Switzerland
Piscataway, NJ: IEEE, 1996
ISBN: 0-7803-3258-X
ISBN: 0-7803-3259-8
ISBN: 0-7803-3260-1
ISBN: 0-7803-3672-0
International Conference on Image Processing (ICIP) <1996, Lausanne>
Fraunhofer IITB ( IOSB) ()
Ablaufplanung; automatic program generation; automatische Programmgenerierung; data flow network; Datenflußnetzwerk; Echtzeit-Bildverarbeitung; Fließbandverarbeitung; flow control planning; machine vision; Maschinensehen; pipeline processing; real-time image processing

High performance vision systems have a complex architecture which makes them difficult to program. The objective of the ARVISA (Advanced Real-time VIsion System Architecture) project, was to facilitate the programming of such systems by providing an integrated application development environment. Its central component is the automatic Flow Control Planner (FCP). Using the FCP, image processing algorithms can be, specified in a device-independent, functional manner. These specifications are then automatically transformed into an executable control program for the desired target device. By examining industrial vision applications, it has been shown that the results of automatic planning by the FCP are comparable to those of manual planning by human experts, but they are created in much less time. FCP thus offers a significant improvement of efficiency for the programming of real-time vision systems.