Abstract
This paper introduces the basic concepts of game theory and outlines the mechanisms for applying game theory models to big data analytics and decision making in the field of geosciences and remote sensing. The author proposes the use of strategic, competitive game theory models for the purpose of spectral band grouping when exploiting hyperspectral imagery. The proposed system uses conflict data filtering based on mutual entropy and a strategy interaction process of multiple band groups in a conflict environment, the goal of which is to maximize the payoff benefit of multiple groups of the whole system. The proposed system uses the Nash equilibrium as the means to find a steady state solution to the band grouping problem, and implements the model under the assumption that all players are rational. The author uses the proposed band grouping as a component in a multi-classifier decision fusion (MCDF) system for automated ground cover classification with hyperspectral imagery. The paper provides experimental results demonstrating that the proposed game theoretic approach significantly outperforms the comparison methods.