Page 324 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

express an object’s assignment to a class. The membership value usually lies between 1.0 and 0.0, where 1.0 expresses a
complete assignment to a class and 0.0 expresses absolutely improbability. The degree of membership depends on the
degree to which the objects fulfil the class-describing conditions. The first step in object-oriented image analysis to form

the processing units by image segmentation. [2] Generally, the following strategies for partitioning a scene into Regions
are distinguished:
- Point-based (Global thresholding)
- Edge-based

- Region-based (split-and merge.). [3]

4-1 THE SEGMENTATION APPROACH

In eCognition, the segmentation approach adopted is multiresolution segmentation, a bottom-up region-growing technique

starting with one-pixel objects. In numerous subsequent steps, smaller image objects are merged into bigger ones.
Throughout this pairwise clustering process, the underlying optimization procedure minimizes the weighted heterogeneity
of resulting image objects. In each step, that pair of adjacent image objects is merged, which stands for the smallest
heterogeneity growth. If the smallest growth exceeds the threshold defined by the scale parameter, the process stops. Doing

so, multiresolution segmentation is a local optimization procedure. [4, 5, 6]
4-2 The role of Multiresolution Segmentation Parameters

The scale parameter is an abstract value with no direct correlation to the object size measured in pixel. It rather depends
on the heterogeneity of the data material.

The scale parameter is related indirectly to the size of the created objects. The heterogeneity at a given scale parameter is
directly linearly dependent on the object size. This way, you will receive objects smaller than in a less heterogeneous image
with the same scale parameter as in a heterogeneous image.
Composition of Homogeneity

The heterogeneity criterion consists of two parts: a criterion for tone and a criterion for shape.
Color/Shape: with these parameters, the influence of color vs. shape homogeneity on the object generation can be adjusted.
The higher the shape criterion, the less spectral homogeneity influences the object generation.
Smoothness/Compactness: when the shape criterion is larger than 0 the user can determine whether the objects shall

become more compact (fringed) or more smooth.
Weight of image channels: this parameter can be used to more or less weight one or more image channels’ influence on
the object generation.
Level: determines whether a newly generated image-level will either overwrite a current level or whether the generated

objects shall contain sub- or super-objects of a still existing level. [8]

3