Objective: Development of algorithms that can search large amounts of imagery for similar spatial, spectral and temporal patterns.
Approach: Our research consists of pattern recognition algorithms using software neural network architectures. The algorithms are tested in the high speed computation and network environment provided by NASA. Within the context of a user browsing a large remote database and searching for specific patterns,a primary goal is minimization of image data transfer over networks by optimal assignment of local workstation and remote parallel computer resources.
Significance: Our work addresses the problem of finding correlated scientific information within very large science databases. The computer tools we are developing will assist the earth scientist in distilling and correlating repeating patterns within the databases.
Status/Plans: Core algorithm finished and tested on serial and parallel (CMP5) computers
Software algorithm submitted to the HPCC/ESS Software Repository and Exchange and made available to the public.
Two presentations of research results at professional conferences, two technical papers published; two manuscripts submitted to professional journals.
Plans include integration of pattern matching system with high speed networks, testing and evaluation of system performance for compressed imagery, and implementation on other high performance computer architectures.
Point of Contact:
Return to the PREVIOUS PAGE