Accession Number:

ADA289872

Title:

Exploiting Parallelism in a Monte Carlo Image-Matching Algorithm.

Descriptive Note:

Final rept. Mar-Jun 94,

Corporate Author:

ARMY RESEARCH LAB ABERDEEN PROVING GROUND MD

Personal Author(s):

• Shires, Dale

Report Date:

1995-01-01

Pagination or Media Count:

60.0

Abstract:

Recovering three-dimensional information from a visual scene requires producing the disparity map of matched stereo images. The matching approach discussed in this paper involves the Metropolis algorithm, a simulated annealing technique. Although simple in principle, this stochastic optimization technique requires immense computational resources. Adapting the algorithm for parallel computing is discussed and demonstrated for several computer architectures.

Descriptors:

• *IMAGE PROCESSING
• *PARALLEL PROCESSING
• *VIDEO MAP MATCHING
• *STEREOSCOPIC RANGE FINDING
• ALGORITHMS
• SIMULATION
• ANNEALING
• OPTIMIZATION
• COMPUTATIONS
• STOCHASTIC PROCESSES
• COMPUTER ARCHITECTURE
• MONTE CARLO METHOD
• THREE DIMENSIONAL
• RESOURCES

Subject Categories:

• Cybernetics
• Optical Detection and Detectors

Distribution Statement:

APPROVED FOR PUBLIC RELEASE