In this chapter, we describe different reconfigurable mesh with buses architectures and show how several image processing problems can be solved efficiently on the weakest of these. The specific problems considered are: area and perimeter of components, shrinking and expanding, clustering, and template matching. In many cases, the resulting algorithms are faster than those for other parallel computer architectures.
|Title of host publication||Machine Intelligence and Pattern Recognition|
|Number of pages||25|
|State||Published - 1 Jan 1994|
|Name||Machine Intelligence and Pattern Recognition|