|
Activity Number:
|
446
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #321423
|
|
|
Title:
|
A Bayesian Hierarchical Data Fusion Approach Leveraging Lidar and Hyperspectral Remote Sensing Information to Improve Aboveground Forest Carbon Estimation for Interior Alaska, USA
|
|
Author(s):
|
Chad Babcock*
|
|
Companies:
|
University of Washington
|
|
Keywords:
|
Gaussian Process ;
Data Fusion ;
Remote Sensing ;
LiDAR ;
Hyperspectral ;
Gaussian Mixture
|
|
Abstract:
|
Multi-sensor data fusion has a long history in remote sensing. The ability to incorporate separate remote sensing sources capable of capturing unique information about a target area is appealing, especially for complex ecological applications. For interior Alaska, merging airborne multi-source remotely-sensed data to model forest carbon can lead to increases in overall and species specific carbon prediction accuracies. I propose to develop a LiDAR and hyperspectral data fusion technique that will effectively combine these disparate data sources in a Bayesian multi-level modeling framework. I envision a Gaussian Process mixture modeling design where mixture weights are modeled using a linear model for spectral unmixing of the hyperspectral data. The finite set of spatial Gaussian Processes model species-specific aboveground forest carbon leveraging forest height information garnered from LiDAR.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2016 program
|
