Image Processing Using a Parallel Architecture.
Abstract
This study developed a set of low level image processing on a parallel computer that allows concurrent processing of images in order to support development of systems that use multiple images to gather information. The parallel computer used is a collection of powerful processors connected in a hypercube topology. The software developed simplifies the interface between the parallel computer and the applications developer by providing a library of functions in the C programming language. These functions are used to control 'image processor programs' that run independently in the parallel computer and perform the image processing operations. The complexities of the parallel processor are hidden and replaced with a flexible structure specifically designed for image processing. This structure provides a simplified interface, but also acts as a framework to which additional image processing operations can be added. Keywords: Image processing, Convolution, Parallel processing.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1987
- Accession Number
- ADA192833
Entities
People
- Billy R. Hodges
Organizations
- Air Force Institute of Technology