Hyperspectral imagery throughput and fusion evaluation over compression and interpolation

Hyperspectral Imagery (HSI) is an emerging capability that extends the analysis of multi-spectral imagery (MSI) through additional bands, variable frequency band distributions, enhanced collections, and improved resolution. These developments have also led to increasing large data files that require intelligent strategies to perform throughput data reduction without degrading exploitation performance. In this paper, we explore the (1) common compression techniques with a novel method that improves the baseline, (2) exploitation targeting with frequency fusion of results over bands to maintain detection, and (3) demonstrate an information fusion performance model strategy for dynamic sensor management of HSI exploitation. The paper describes a method for robust HSI performance evaluation to truncate disturbances, interpolate data across these locations, compress and reconstruct the signal, perform decision-fusion detection, and check the error associated with these operations - all supporting techniques to enable realizable HSI tracking and identification solutions.