Algorithms in computer vision are computationally intensive and have distin
ct characteristics. Message-passing MIMD computers are popular architecture
s for computer vision due to their flexibility and asynchronous structure.
However, performance of parallel programs executed on these systems is depe
ndent on the efficiency of inter-processor communication. This paper descri
bes a software system developed to handle complex communication mechanisms
for a system designed for computer vision applications. The system is const
ructed with TI320C40 DSPs and employs the distributed memory MIMD paradigm.
A set of experiments has been conducted to determine the hardware paramete
rs of the communication links of the processors. Furthermore, the performan
ces of the communication algorithms developed, which are frequently require
d in parallel vision algorithms, are presented. These results will be usefu
l in optimizing performance of parallel programs developed for the system.
(C) 2000 Published by Elsevier Science B.V. All rights reserved.