Interactive Pansharpening and Active Classification in Remote Sensing

This paper presents two multimodal prototypes for remote sensing image classification where user interaction is an important part of the system. The first one applies pansharpening techniques to fuse information from different satellite sensors to obtain a high resolution (HR) multispectral image. Once the HR image has been classified the user can interact with the system to select a class of interest. The pansharpening parameters are then modified to increase the system accuracy for the selected class without deteriorating the performance of the classifier on the other classes. The second prototype utilizes Bayesian modeling and inference to implement active learning and parameter estimation in binary kernel-based multispectral classification problems with possibly infinite dimensional feature spaces. In the prototype we developed three different strategies for selecting the more informative pixel to be included in the training set. In the experimental section, the prototypes are described and applied to two real multispectral image classification problems.