defOn Supporting Rule-Based Image Interpretation Using a Distributed Memory Multicomputer

High-level tasks in image understanding are similar to many AI problems in terms of complexity and solution techniques. In particular, rule based systems have been employed for image interpretation using domain knowledge. Past research efforts on the hardware support of rule-based systems typically assume that little data access locality can be explored. Thus, the Rete network is widely used, and a shared-memory architecture is often employed for supporting such systems. However, such assumptions may not be true for image interpretation. Given a suitable image partitioning and mapping strategy that results in significant data locality, coarse grain parallelism can be fruitfully exploited on a distributed memory multicomputer. This chapter studies these issues through the parallel implementation of a rule-based system for image interpretation. The implementation is done on an AT&T PIXEL machine, which employs a distributed memory computer architecture with message-passing. The merits an..