A fundamental challenge for parallel computing is to obtain high-level, architecture independent, algorithms which execute efficiently on general-purpose parallel machines.With ing problem posed by the NAS Integer Sorting ( 1S) Benchmark.

Develops efficient algorithms for low and intermediate level image processing on the scan line array processor, a SIMD machine consisting of a linear array of cells that processes images in a scan line fashion. For low level processing, the authors present algorithms for block DFT, block DCT, convolution, template matching, shrinking, and expanding which run in real-time. By real-time, the authors mean that, if the required processing is based on neighborhoods of size m/spl times/m, then the output lines are generated at a rate of O(m) operations per line and a latency of O(m) scan lines, which is the best that can be achieved on this model. The authors also develop an algorithm for median filtering which runs in almost real-time at a cost of O(m log m) time per scan line and a latency of [m/2] scan lines. For intermediate level processing, the authors present optimal algorithms for translation, histogram computation, scaling, and rotation. The authors also develop efficient algorithms for labelling the connected components and determining the convex hulls of multiple figures which run in O(n log n) and O(n log/sup 2/n) time, respectively. The latter algorithms are significantly simpler and easier to implement than those already reported in the literature for linear arrays.>

We develop efficient algorithms for low and intermediate level image processing on the scan line array processor that handles images in a scan line fashion. For low level processing, we present algorithms for block DFT, block DCT, convolution, template matching, shrinking, and expanding. These algorithms run in real-time - that is, the output lines are generated at the rate of O(m) time per line, where the required processing is based on neighborhoods of size m x m. For intermediate level processing, we present efficient algorithms for scaling, translation, connected components, and convex hulls of multiple figures.