Uncertainty Gated Network for Land Cover Segmentation
This addresses the open problem of land cover segmentation in remote sensing, which is incremental as it builds on existing deep learning methods.
The paper tackles the problem of modeling multiscale contexts for land cover segmentation in remote sensing by proposing a novel approach that uses heteroscedastic uncertainty to gate network layers, resulting in improved pixel classification decisions.
The production of thematic maps depicting land cover is one of the most common applications of remote sensing. To this end, several semantic segmentation approaches, based on deep learning, have been proposed in the literature, but land cover segmentation is still considered an open problem due to some specific problems related to remote sensing imaging. In this paper we propose a novel approach to deal with the problem of modelling multiscale contexts surrounding pixels of different land cover categories. The approach leverages the computation of a heteroscedastic measure of uncertainty when classifying individual pixels in an image. This classification uncertainty measure is used to define a set of memory gates between layers that allow a principled method to select the optimal decision for each pixel.