Binary Erosion-Reconstruction

Author: HfT Stuttgart (c) 2013
Menu: Grid | Filter

Description

Common binary opening does not guarantee, that foreground regions which outlast the erosion step are reconstructed to their original shape in the dilation step. Depending on the application, that might be considered as a deficiency. Therefore this tool provides a combination of erosion with the binary Geodesic Morphological Reconstruction, see Vincent (1993). Here we use the algorithm on p. 194: Breadth-first Scanning. The marker is defined as the eroded input grid, whereas the mask is just the input grid itself. The output grid is the reconstruction of the marker under the mask.

References

Arefi, H., Hahn, M. (2005): A Morphological Reconstruction Algorithm for Separating Off-Terrain Points from Terrain Points in Laser Scanning Data. Proceedings of the ISPRS Workshop Laser Scanning 2005, Enschede, the Netherlands, September 12-14, 2005 PDF.
Vincent, L. (1993): Morphological Grayscale Reconstruction in Image Analysis: Applications and Efficient Algorithms. IEEE Transactions on Image Processing, Vol. 2, No 2. doi: 10.1109/83.217222.

Parameters

	Name	Type	Identifier	Description	Constraints
Input	Input Grid	grid, input	`INPUT_GRID`	Grid to be filtered	-
Output	Output Grid	grid, output	`OUTPUT_GRID`	Reconstruction result	-
Options	Grid System	grid system	`PARAMETERS_GRID_SYSTEM`	-	-
Options	Filter Size (Radius)	integer number	`RADIUS`	Filter size (radius in grid cells)	Default: 3

Command Line


Usage: saga_cmd grid_filter 13 [-INPUT_GRID ] [-OUTPUT_GRID ] [-RADIUS ]
  -INPUT_GRID: 	Input Grid
	grid, input
  -OUTPUT_GRID:	Output Grid
	grid, output
  -RADIUS:     	Filter Size (Radius)
	integer number
	Default: 3