Nate Currit

Automatic ground filtering for Light Detection And Ranging (LIDAR) data is a critical process for Digital Terrain Model (DTM) and three-dimensional urban model generation. Although researchers have developed many methods to separate bare ground from other urban features, the problem has not been fully solved due to the similar characteristics possessed by ground and non-ground objects, especially on abrupt surfaces.

Extracting water bodies from remotely sensed imagery is an important procedure for many water related studies. It is considered as a challenging problem of automatically and effectively extracting water bodies from remotely sensed images, and many researchers rely on existing Geoinformation System (GIS) data or manual digitizing to obtain water body boundaries. This work firstly generalizes several water body characteristics in optical remotely sensed images and then proposes a segmentation-based water body extraction method. The proposed algorithm exploits several features and uses a perception machine (PM) of neural network (NN) to build a classifier. An overall classification accuracy of 96% indicates that this method holds promise for extracting water bodies from optical remotely sensed images.