论文信息 - Mapping an N N Image on P Processors

Mapping an N N Image on P Processors

An cvaluation is made of hardware and software rcquircmcnts to support window-, crinkle- and pyramid-mapping on a lincar and a squarc proccssor array. The diffcrcnt mapping stratcgics also lcad to diffcrcnt algorithms to solvc imagc proccssing problcms. Thrcc imagc proccssing problems servc as an cxarnplc of thc algorithmic diffcrcnccs causcd by the mapping mcthods. For thc invcstigatcd 2D proccssor array, thc pyramid mapping givcs thc fastcst rcsults.

[1] D. H. Schaefer,et al. The GAM pyramid , 1987 .

[2] Steven L. Tanimoto,et al. Two Pyramid Machine Algorithms for Sdge Detection in Noisy Binary Images , 1983, Inf. Process. Lett..

[3] E. R. Komen,et al. Low-level image processing architectures: Compared for some non-linear recursive neighbourhood operations , 1990 .

[4] Robert P. W. Duin,et al. Implementation and Use of Software Scanning on a Small CLIP4 Processor Array , 1988, Pattern Recognition.

[5] Anthony P. Reeves,et al. Parallel computer architectures for image processing , 1984, Comput. Vis. Graph. Image Process..

[6] Stefano Levialdi,et al. Pyramidal Systems for Computer Vision , 2011, NATO ASI Series.

[7] Kenneth E. Batcher,et al. Design of a Massively Parallel Processor , 1980, IEEE Transactions on Computers.

[8] Alexander A. Sawchuk,et al. Binary image algebra and optical cellular logic processor design , 1989, Comput. Vis. Graph. Image Process..

[9] Leonard Uhr,et al. Parallel computer vision , 1987 .