Design and evaluation of a grassfire skeletonization chip
B. Oláh, S. Török, A. Poppe, P. Masa, and A. Lörincz
In: Proceedings of Conf. MIXDES99, Poland, July, 1999
Abstract
We present a novel mixed-signal VLSI focal-plane array processor performing
the morphological operation of skeletonization of binary images at a processing
speed of up to 10K frames/second. The chip exploits the inherent parallelism of
the grassfire algorithm by using a massively parallel mixed-mode array achieving
high computational speed and low power consumption. The system is a two-dimensional
array of identical processing elements (PEs) with integrated photodetectors allowing
parallel optical input. The skeleton computation is performed in a fully parallel,
asynchronous manner by local interactions between PEs, the transformed image is
represented by analog values. The output of the transformation can be serially
accessed with an on-chip raster-scan circuit generating multiscanning-compatible
video signal. The system was physically designed and prototypes of 22x26 array
were fabricated in a standard 0.7 micron CMOS process. Circuit and layout design
considerations as well as output data samples of the fully functional fabricated
chip are presented.
