Multichannel integration for landcover classification in satellite imagery

This paper presents a methodology and results for multichannel integration in remotely sensed data by learning disparate models from each channel for information classification. The objective is the classification of data to map landcover. The methodology is based on a modular structure consisting of multiple classifiers, each of which solves the problem independently based on its input observations. Each classifier module is trained to detect distinct landcover regions and a higher order decision integrator collects evidence from each of the modules to delineate a final region. A Bayesian realization of the framework is developed, where each classifier module represents the conditional probability density function. Results of classification are shown in Landsat data. These integrated results are also compared to single-channel/feature classification results.