A high-speed implementation of manifold coordinate representationsof hyperspectral imagery: A GPU-based approach to rapid nonlinearmodeling

摘要

Working with the Naval Research Laboratory, Celestech has implemented advanced non-linear hyperspectral image(HSI) processing algorithms optimized for Graphics Processing Units (GPU). These algorithms have demonstratedperformance improvements of nearly 2 orders of magnitude over optimal CPU-based implementations. The paper brieflycovers the architecture of the NIVIDIA GPU to provide a basis for discussing GPU optimization challenges andstrategies. The paper then covers optimization approaches employed to extract performance from the GPUimplementation of Dr. Bachmann's algorithms including memory utilization and process thread optimizationconsiderations.The paper goes on to discuss strategies for deploying GPU-enabled servers into enterprise service oriented architectures.Also discussed are Celestech's on-going work in the area of middleware frameworks to provide an optimized multi-GPUutilization and scheduling approach that supports both multiple GPUs in a single computer as well as across multiplecomputers.This paper is a complementary work to the paper submitted by Dr. Charles Bachmann entitled "A Scalable Approach toModeling Nonlinear Structure in Hyperspectral Imagery and Other High-Dimensional Data Using Manifold CoordinateRepresentations". Dr. Bachmann's paper covers the algorithmic and theoretical basis for the HSI processing approach.