A VLSI architecture is proposed for the realization of real-time two-dimens
ional (2-D) image filtering in an address-event-representation (AER) vision
system. The architecture is capable of implementing any convolutional kern
el F(x,y) as long as it is decomposable into x-axis and y-axis components,
i.e., F(x, y) = H(x)V(y), for some rotated coordinate system {x, y} and if
this product can he approximated safely by a signed minimum operation. The
proposed architecture is intended to be used in a complete vision system, k
nown as the boundary contour system and feature contour system (BCS-FCS) vi
sion model, proposed by Grossberg and collaborators. The present paper prop
oses the architecture, provides a circuit implementation using MOS transist
ors operated in weak inversion, and shows behavioral simulation results at
the system level operation and some electrical simulations.