Tool Name	Line Smoothing
Tool ID	7
Library ID	shapes_lines
Version	1.0
Author(s)	V. Wichmann, J. Engels (c) 2014

Description

The tool provides methods for line smoothing including iterative averaging (SIA) and Gaussian filtering.

Iterative averaging (SIA) is described by Mansouryar & Hedayati (2012). A higher smoothing sensitivity results in a stronger smoothing in less iterations and vice versa. The 'improved SIA model' simply applies a preservation factor in the first iteration and then runs the 'basic SIA model' for the following iterations.

Gaussian filtering with shrinkage correction is described by Lowe (1989).

In case the density of line vertices is too high, the 'Line Simplification' tool can be applied first. If the density of line vertices is too low, additional vertices can be inserted by applying the 'Convert Lines to Points' and the 'Convert Points to Line(s)' tools prior to smoothing.

References

Lowe, D. (1989): Organization of Smooth Image Curves at Multiple Scales. International Journal of Computer Vision, 3, 119-130. PDF

Mansouryar, M., Hedayati, A. (2012): Smoothing Via Iterative Averaging (SIA) - A Basic Technique for Line Smoothing. International Journal of Computer and Electrical Engineering, Vol. 4, No. 3, 307-311.

Parameters

Name	Type	Identifier	Description	Constraints
Input
Lines	shapes, input	LINES_IN	The input line shapefile to smooth.
Output
Smoothed Lines	shapes, output	LINES_OUT	The smoothed output line shapefile.
Options
Method	choice	METHOD	Choose the method to apply.	Available Choices: [0] basic SIA model [1] improved SIA model [2] Gaussian Filtering Default: 2
Sensitivity	integer number	SENSITIVITY	Half the size of the moving window [vertex count], controls smoothing sensitivity.	Minimum: 1 Default: 3
Iterations	integer number	ITERATIONS	The number of smoothing iterations [-].	Minimum: 1 Default: 10
Preservation	floating point number	PRESERVATION	The smoothing preservation factor [-].	Minimum: 1.000000 Default: 10.000000
Sigma	floating point number	SIGMA	Standard deviation of the Gaussian filter [-].	Minimum: 0.500000 Default: 2.000000