PARALLEL ALGORITHMS FOR IMAGE-ENHANCEMENT AND SEGMENTATION BY REGION GROWING, WITH AN EXPERIMENTAL-STUDY

This paper presents efficient and portable implementations of a powerf ul image enhancement process, the Symmetric Neighborhood Filter (SNF), and an image segmentation technique that makes use of the SNF and a v ariant of the conventional connected components algorithm which we cal l delta-Connected Components. We use efficient techniques for distribu ting and coalescing data as well as efficient combinations of task and data parallelism. The image segmentation algorithm makes use of an ef ficient connected components algorithm based on a novel approach for p arallel merging. The algorithms have been coded in SPLIT-C and run on a variety of platforms, including the Thinking Machines CM-5, IBM SP-1 and SP-2, Gray Research T3D, Meiko Scientific CS-2, Intel Paragon, an d workstation clusters. Our experimental results are consistent with t he theoretical analysis (and provide the best known execution times fo r segmentation, even when compared with machine-specific implementatio ns). Our test data include difficult images from the Landsat Thematic Mapper (TM) satellite data.