Advancing Hyperspectral Signature Integration with Airborne and Ground-based Laser Technology

Combining high spatial resolution data with high spectral resolution imagery to improve classification accuracy in remotely sensed data is the goal of this effort. This requires 1) an instrument or instruments capable of acquiring this datacoincidentally, and 2) proper registration, pre-processing, classification and evaluation algorithms to determine the best way to combine the data to achieve the most significant improvement in classification accuracy.Surface Optics Corporation (SOC) will use its hyperspectral imager, the SOC-700, in conjunction with the I-Site 3D Laser Imager to collect spatial and spectral data from various regions, then use this data to explore the most effective means of combiningthe data to improve classification accuracy. This includes the use of derived measures from both the spatial and spectral data. Evaluation of effectiveness will be done with standard classification accuracy measures. The result of the study will be aranking of the various techniques and data measures that were undertaken in the study in terms of their effectiveness, with the eventual goal being to incorporate them as part of a complete Lidar/Hyperspectral data acquisition system. Benefits of aHyperspectral/Lidar imaging system would be broad based and dual-use, providing the capability for improved resource management for civil communities by constructing more detailed census of vegetation coverage and assisting in the environmental managementof federal, state and local lands in sensitive habitats.