Feature-Based Transformers

A Feature-Based Transformer is one which processes only a single feature, one at a time and in isolation from all other features; for example, a LengthCalculator. The act of processing in this way is known as Feature-Based Processing.

Some examples of Feature-Based restructuring are:

• Measurements – length and area calculations are performed on only one feature at a time, and the area of one feature has no impact on the area of another.
• Line Generalization – each line feature is generalized in turn with no reference to surrounding features.
• Center of Gravity Calculations – FME can calculate the “center of gravity” (the geographic center) of an area feature. Each calculation is unique and independent of other features.

In general, most transformers dealing with attribute data are feature-based, while spatial data handling transformers are mostly Group-Based Transformers.

Categorization

Categorizing feature-based transformers is usually straightforward, the general definition being that features are processed in isolation. When a single feature is processed by itself, but in relation to other features (for example the Snapper), then this is usually classed as a Group-Based process since the features have to be held together in memory to achieve this.

Flow of Features

Feature-based transformers follow the general rule of FME, that features are processed one at a time.