Generalizer

Simplifies, smoothes, measures, or fits geometry according to a variety of algorithms.

Jump to Configuration

Typical Uses

Simplifying geometry for use at different scales
Simplifying geometry to optimize storage or processing time
Smoothing jagged lines

How does it work?

The Generalizer receives features with vector geometry and applies a selected algorithm.

There are four types available, and different algorithms may be more suitable for particular purposes or datasets:

Generalize: These algorithms simplify features by removing vertices.
- Douglas and Douglas with Arc Fitting - Remove vertices that deviate little.
- Thin and Thin No Point - Remove vertices that are close together.
- Deveau - Remove vertices with less effect on shape.
- Wang - Remove insignificant bends.
Smooth: These algorithms move vertices to produce smoother geometry.
- McMaster and McMaster Weighted Distance - Move vertices based on their neighbors.
- NURBfit - Fit smooth curves to the geometry and stroke them.
- Fillet - Round sharp corners.
Measure: This algorithm finds inflection points.
- Inflection Points - Produce points to measure sinuosity.
Fit: This algorithm produces a two-point line based on vertices.
- Orthogonal Distance Regression - Find an optimal path through vertices.

Areas that have common boundaries may have those shared portions optionally preserved.

Generalizing Algorithms

Douglas and Douglas With Arc Fitting

The Douglas algorithm removes vertices that deviate less than a specified distance from a line connecting vertices within the geometry.

The vertices are evaluated by initially connecting the first and last point, identifying vertices to be kept, then recursively evaluating shorter and shorter segments until only points identified as kept remain.

Critical points are retained and vertices are not moved. When applied to a polygon, the first point will always be kept.

Douglas with Arc Fitting also fits arcs to the geometry where possible to further reduce the number of vertices. Output features may have path geometry.

Based on the Ramer–Douglas–Peucker algorithm.

Input Feature
Douglas
Douglas With Arc Fitting

Algorithm Parameters

Generalization Tolerance

Maximum deviation in ground units. Vertices offset less than this distance are discarded.

Thin and Thin No Point

The Thin algorithm removes vertices that are less than a specified distance from an adjacent vertex. First and last points are not removed.

If a feature’s entire length is less than the Generalization Tolerance, it is replaced either by a point at the last coordinate (Thin) or a two-point line connecting the first and last vertices (Thin No Point).

Input Feature

Thinned

Algorithm Parameters

Generalization Tolerance

Maximum distance between vertices in ground units.

Deveau

The Deveau algorithm removes vertices that contribute less to the overall shape of the feature. It uses rectangular bands of tolerance that move along the feature to determine which vertices are discarded, also evaluating angles to identify significant features to keep.

Smoothness can be increased to further reduce vertices.

Output features are 2D and any measures are removed.

Based on the Deveau algorithm.

Input Feature

Deveau

Minimum Smoothness (1)

Deveau

Maximum Smoothness (30)

Algorithm Parameters

Generalization Tolerance

Width of the tolerance band in ground units. A wider band removes more vertices.

Corresponds to Deveau’s TOL variable.

Smoothness Factor

An integer ranging from 1 to 30. Higher smoothness values remove more vertices.

Default: 1

Corresponds to Deveau’s NW variable.

Sharpness Angle

Spikes with a vertex angle less than this value and height and width greater than the Generalization Tolerance are considered significant features and will not be blunted.

Accepted values: 0 to 180.0 degrees

Default: 110

Corresponds to Deveau’s ANG variable.

Wang

The Wang algorithm removes insignificant bends by iteratively evaluating them against a reference circle.

For each bend, a value is calculated - 75% of the area of a circle with circumference equal to the bend’s perimeter. If this value is less than the area of the reference circle, the bend is generalized.

Based on the Wang-Müller algorithm.

Algorithm Parameters

Generalization Tolerance

Diameter of the reference circle.

Smoothing Algorithms

McMaster and McMaster Weighted Distance

The McMaster algorithms smooth lines by adjusting vertex position based on neighboring vertices.

The coordinates of a specified number of neighbors on both sides of the vertex are averaged to produce a new location. The vertex is moved a specified percentage of the distance between the new and original locations, with the default of 50 being halfway between. Lower percentages move nearer the new location and higher percentages move nearer the original.

When using weighted distance, inverse distance weighting is applied before averaging. Weight is the reciprocal of the distance from the vertex being smoothed multiplied by the Weighting Power value. Neighbors that are further away have less effect than those nearby.

The overall effect using either method is to pull vertices towards their neighbors, smoothing the line.

Geometry is not simplified - output features have the same number of vertices as input.

The first and last vertices of features are handled according to the Number of Neighbors (N):

Lines: The first N and last N vertices are not smoothed.
Areas: All vertices are smoothed.
Adjacent Areas: When Preserve Shared Boundaries is Yes, collinear portions of their boundaries will be smoothed together. Unshared portions are smoothed as lines (first N and last N vertices are not smoothed).

Based on McMaster's Sliding Averaging Algorithm and McMaster's Distance-Weighting Algorithm.

Input Features

McMaster

McMaster Weighted Distance

Algorithm Parameters

Number of Neighbors

The number of points on either side of the vertex to include in the averaged location.

Default: 2

Displacement Percentage

The final position of the vertex as a percentage of the distance between the averaged location and original location.

Default: 50

Weighting Power

When using McMaster Weighted Distance, a value to multiply the vertex distance by.

Default: 1.0

NURBfit

The NURBfit algorithm fits B-Spline curves to the input features and strokes the results to output linear geometry.

Spline shapes are determined by the polynomial degree and desired output segment length. Higher degrees produce smoother curves. Segment Length is measured in ground units and controls the density of vertices on the output feature.

See Non-uniform rational B-spline

Input Features

NURBfit

Algorithm Parameters

Degree of Basis Polynomial

Degree of the polynomial used to approximate the curve. The higher the degree, the smoother the line.

Accepted values:Integers >= 2

Default: 2

Segment Length

Output segment length in ground units.

If zero (0), the segment length will be calculated so that the output feature has 10x the number of vertices as the input feature.

Accepted values: Numbers >= 0

Default: 0

Fillet

The Fillet algorithm replaces corners with arcs.

If an arc of the specified radius does not fit in a corner, it is output unchanged.

Input Features

Fillet

Algorithm Parameters

Fillet Radius

Arc radius in ground units.

Accepted values: Numbers > 0

Measuring Algorithm

Inflection Points

The Inflection Points algorithm identifies a feature’s inflection points and outputs them as aggregate multi-point geometry.

The calculation incorporates neighboring vertices, which can be adjusted. Zero (0) means no neighbors are considered. Increasing the number of neighbors smooths the line and may result in fewer inflection points.

See Sinuosity and Inflection Points.

Input Feature

Inflection Points

Algorithm Parameters

Number of Neighbors

The number of adjacent vertices to consider in inflection point calculation.

Accepted values: Integers >= 0

Default: 2

Fitting Algorithm

Orthogonal Distance Regression

The Orthogonal Distance Regression algorithm replaces the feature's geometry with a two-point line that minimizes the orthogonal distance between it and the original geometry's vertices.

Orthogonal distance is the shortest perpendicular distance between a point and a line.

Input Feature
Orthogonal Distance Regression

Algorithm Parameters

This algorithm has no parameters.

Usage Notes

To maintain topologies that involve other features while generalizing, consider using the SherbendGeneralizer.
The Densifier may be useful in cases where generalization is too aggressive in portions of the data.

Configuration

Input Ports

Output Ports

Parameters

Group Processing

Group By

The default behavior is to use the entire set of input features as the group. This option allows you to select attributes that define which groups to form. Each set of features which have the same value for all of these attributes will be processed as an independent group.

Complete Groups

Select the point in processing at which groups are processed:

When All Features Received	This is the default behavior. Processing will only occur in this transformer once all input is present.
When Group Changes (Advanced)	This transformer will process input groups in order. Changes of the value of the Group By parameter on the input stream will trigger processing on the currently accumulating group. This may improve overall speed (particularly with multiple, equally-sized groups), but could cause undesired behavior if input groups are not truly ordered.

General

Preserve Shared Boundaries

Select a method for handling adjacent areas that have boundaries in common:

No (default): Features are processed individually without considering neighboring features. May produce gaps and overlaps.
Yes: Shared boundaries are processed as one by computing topology, performing generalization, then recreating areas.

Note that overlapping areas may need pre-processing with an AreaOnAreaOverlayer or Snapper(Vertex Snapping) for expected results.

Shared Boundaries Tolerance

The minimum distance between boundaries in 2D before they are considered shared, in ground units. If the tolerance is Automatic, a tolerance will be automatically computed based on the location of the input geometries.

Preserve Path Segments

Select a method for handling features with path geometry:

No: Path segments may be joined to form a new line.
Yes: Path segments will not be joined to form a new line.

Editing Transformer Parameters

Transformer parameters can be set by directly entering values, using expressions, or referencing other elements in the workspace such as attribute values or user parameters. Various editors and context menus are available to assist. To see what is available, click beside the applicable parameter.

How to Set Parameter Values

Defining Values

There are several ways to define a value for use in a Transformer. The simplest is to simply type in a value or string, which can include functions of various types such as attribute references, math and string functions, and workspace parameters.

Using the Text Editor

The Text Editor provides a convenient way to construct text strings (including regular expressions) from various data sources, such as attributes, parameters, and constants, where the result is used directly inside a parameter.

Text Editor

Using the Arithmetic Editor

The Arithmetic Editor provides a convenient way to construct math expressions from various data sources, such as attributes, parameters, and feature functions, where the result is used directly inside a parameter.

Arithmetic Editor

Conditional Values

Set values depending on one or more test conditions that either pass or fail.

Parameter Condition Definition Dialog

Content

Expressions and strings can include a number of functions, characters, parameters, and more.

When setting values - whether entered directly in a parameter or constructed using one of the editors - strings and expressions containing String, Math, Date/Time or FME Feature Functions will have those functions evaluated. Therefore, the names of these functions (in the form @<function_name>) should not be used as literal string values.

Content Types

String Functions	These functions manipulate and format strings.
Special Characters	A set of control characters is available in the Text Editor.
Math Functions	Math functions are available in both editors.
Date/Time Functions	Date and time functions are available in the Text Editor.
Math Operators	These operators are available in the Arithmetic Editor.
FME Feature Functions	These return primarily feature-specific values.
FME Parameters	FME and workspace-specific parameters may be used.
Creating and Modifying User Parameters	Create your own editable parameters.

Dialog Options - Tables

Table Tools

Transformers with table-style parameters have additional tools for populating and manipulating values.

Row Reordering

Enabled once you have clicked on a row item. Choices include:

Add a row
Remove a row
Move current row up one
Move current row down one
Move current row to top
Move current row to bottom

Cut, Copy, and Paste

Enabled once you have clicked on a row item. Choices include:

Cut a row - delete and copy to clipboard
Copy a row to the clipboard
Paste a row from the clipboard

Cut, copy, and paste may be used within a transformer, or between transformers.

Filter

Start typing a string, and the matrix will only display rows matching those characters. Searches all columns. This only affects the display of attributes within the transformer - it does not alter which attributes are output.

Import

Import populates the table with a set of new attributes read from a dataset. Specific application varies between transformers.

Reset/Refresh

Generally resets the table to its initial state, and may provide additional options to remove invalid entries. Behavior varies between transformers.

Note: Not all tools are available in all transformers.

For more information, see Transformer Parameter Menu Options.

Reference

Processing Behavior	Feature-Based
Feature Holding	No
Dependencies	None
Aliases	LineGeneralizer AreaGeneralizer AreaSmoother LineSmoother
History

FME Community

The FME Community has a wealth of FME knowledge with over 20,000 active members worldwide. Get help with FME, share knowledge, and connect with users globally.

Search for all results about the Generalizer on the FME Community.

Examples may contain information licensed under the Open Government Licence – Vancouver, Open Government Licence - British Columbia, and/or Open Government Licence – Canada.

Number of Neighbors	See McMaster and McMaster Weighted Distance or Inflection Points.
Displacement Percentage	See McMaster and McMaster Weighted Distance
Weighting Power	See McMaster and McMaster Weighted Distance

Degree of Basis Polynomial	See NURBfit
Segment Length	See NURBfit