Building a Multi-Processor, Multi-Threaded, Multi-Resolution Bathymetric Data Processing Algorithm

Brian Calder
Research Associate Professor


Monday, Mar. 7, 2011, 11:20am
Chase 130

With bathymetric data capture rates now routinely on the order of tens of millions of soundings per hour and total survey data volumes in the billions of soundings, even current best-of-breed computer-assisted hydrography algorithms are destined eventually to be outpaced. If in addition we consider the problem of extreme dynamic depth range within one survey and the requirements for high-resolution bathymetric models to maintain the best-available representation of the seafloor, then in addition to computational complexity, the problem becomes one of appropriate representation and efficient data access.This presentation considers the design and implementation of a data processing algorithm that attempts to address the problem of appropriate data representation in a computationally efficient manner with particular emphasis on the hydrographic data processing problem. The algorithm, derived from the CUBE core estimation scheme, can scale to almost arbitrarily large geographic areas, adapt to different data representation resolutions based on metrics derived from the data-as-captured, and has a flexible, multi-threaded, multi-processor capable client-server implementation. We consider the problems of stable data-driven estimation, efficient data management, and the extent to which research-based code can, or should, be the basis for a potential transition to commercial implementation. In the process we examine the uses of the confluent hypergeometric function, the potential perils of dyadic sampling, and the proper time to pick blueberries.


Brian Calder has a Ph.D in Computing and Electrical Engineering, completing his thesis on Bayesian methods in Sidescan Sonar processing in 1997. Since then he has worked on a number of signal processing problems, including real-time grain size analysis, seismic processing, and wave-field modeling for shallow seismic applications.

His research interests include methods for error modeling, propagation and visualization, and adaptive sonar backscatter modeling. His work has focused on developing methods for textural analysis of seafloor sonar data, as well as exploring innovative approaches to target detection and seafloor property extraction.

Dr. Calder is currently focusing on statistically robust automated data cleaning approaches and tracing uncertainty in hydrographic data.