A. Bugeja et Ww. Yang, A RECONFIGURABLE VLSI COPROCESSING SYSTEM FOR THE BLOCK MATCHING ALGORITHM, IEEE transactions on very large scale integration (VLSI) systems, 5(3), 1997, pp. 329-337
Several VLSI architectures for the full-search block matching algorith
m have been proposed in recent years due to its computation and I/O-in
tensive nature and its importance in various computer vision and image
processing applications, This paper presents a new coarse grained rec
onfigurable coprocessor which is suitable for integration with general
purpose microprocessors. The 180 000 transistor custom VLSI design wa
s implemented in 0.6 mu m CMOS on a 4.12 mm x 2.59 mm die and has been
fully tested up to 33 MHz. For a typical image database search applic
ation, a sample system consisting of four coprocessors interfaced thro
ugh a 33 MHz PCI bus will provide a speedup of 320x over an 80486 DX2/
66 MHz and 64x over a 150-MHz Pentium running fully optimized assembly
code.