Paper
17 March 1983 Neighborhood Access Problem In Image Parallel Machines
Per-Erik Danielsson
Author Affiliations +
Abstract
Image parallelism implies a SIMD-architecture where a two-dimensional array of processors and memory modules are controlled by a central master control unit. It is emphasized that effective access of artribrarily sized neighborhoods (which is everso common in image processors) are distributed over the image plane. The access and interconnection problem for such an architecture is shown to have several possible solutions with different trade-offs. No existing design presents good solutions of the neighborhood access problem. It is shown that a proper design in this respect results in a 13-fold increase in speed compared to what has recently been reported for the MPP.
© (1983) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Per-Erik Danielsson "Neighborhood Access Problem In Image Parallel Machines", Proc. SPIE 0359, Applications of Digital Image Processing IV, (17 March 1983); https://doi.org/10.1117/12.965953
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image processing

Convolution

Digital image processing

Array processing

Nonlinear filtering

Data modeling

Linear filtering

RELATED CONTENT

Architecture for processing image algebra operations
Proceedings of SPIE (June 01 1992)
A High-Speed Processor for Large Imagery
Proceedings of SPIE (April 05 1989)
Image smoothing with Savitzky-Golay filters
Proceedings of SPIE (May 30 2003)
Malleable architecture generator for FPGA computing
Proceedings of SPIE (October 21 1996)
A Relaxation Based Adaptive Filtering Algorithm
Proceedings of SPIE (December 04 1984)
Parallel programming for the parallel recirculating pipeline
Proceedings of SPIE (November 01 1990)

Back to Top