Evaluation of Current Filtering Algorithms
Ahmad Fikri Abdullah · 2020
The main assumption of this algorithm is that points in a given height range adjacent to a particular measurement point are for bare earth. In the 1D Progressive Morphological algorithm the lowest point in a neighbourhood is labelled as a terrain point. The 2D Progressive Morphological algorithm adopts the same concept. The only difference is that the filtering window is a two-dimensional rectangle or any other shape. The main assumption of this algorithm is that the elevation differences of neighbouring measurement points are distinct between ground, trees, and buildings, in a limited sized area within a search window. The assumption of this algorithm is that the lowest point in a set of neighbouring measurement points belongs to the ground. The main assumption of this algorithm is that the ground is essentially represented as being continuous, or at least, a piecewise continuous surface. A polynomial 2-surface filter uses a mathematical function to approximate the ground.