STUDIES IN PARALLEL IMAGE PROCESSING (ARCHITECTURE)

GIE-MING LIN, Purdue University

Abstract

The supervised relaxation operator combines the information from multiple ancillary data sources with the information from multispectral remote sensing image data and spatial context. Iterative calculations integrate information from the various sources, reaching a balance in consistency between these sources of information. The supervised relaxation operator is shown to produce substantial improvements in classification accuracy compared to the accuracy produced by the conventional maximum likelihood classifier using spectral data only. The convergence property of the supervised relaxation algorithm is also described. Improvement in classification accuracy by means of supervised relaxation comes at a high price in terms of computation. In order to overcome the computation-intensive problem, a distributed/parallel implementation is adopted to take advantage of a high degree of inherent parallelism in the algorithm. To accomplish this, first, a graphical modeling and analysis method is described for algorithms implemented using the SIMD mode of parallelism. Second, the comparison of execution times between SIMD and MIMD modes is discussed, based on the analysis of implicit and explicit operations embedded in an algorithm. From the comparison it is shown that some algorithms are suitable for MIMD mode of parallelism; some are more suited for SIMD. Third, several performance measures for SIMD, functions of problem size and system size, are discussed. Finally, two multistage interconnection networks to support the distributed/parallel system are overviewed. Based on the these considerations, an optimal system configuration in terms of execution time and system cost is proposed.

Degree

Ph.D.

Subject Area

Electrical engineering

COinS