1. UZAKTAN ALGIILAMA UYGULAMALARI Segmentasyon Algoritmaları
Prof. Dr. Oğuz Güngör
Karadeniz Teknik Üniversitesi
Harita Mühendisliği Bölümü
61080 Trabzon

2. Image Layer Weights
Image layers can be weighted depending on their importance or suitability for the segmentation result.
The higher the weight assigned to an image layer, the more weight will be assigned to that layer’s pixel information during the segmentation process, if it is used.
Consequently, image layers that do not contain the information intended for representation by the image objects should be given little or no weight.
For example, when segmenting a geographical LANDSAT scene using multiresolution segmentation or spectral difference segmentation, the segmentation weight for the spatially coarser thermal layer should be set to 0 in order to avoid deterioration of the segmentation result by the blurred transient between image objects of this layer.

3. Thematic Layer Weights
In the Thematic Layers field, specify the thematic layers to be considered in addition to segmentation.
Each thematic layer used for segmentation will cause additional splitting of image objects while enabling consistent access to its thematic information.
You can segment an image using more than one thematic layer.
The results are image objects representing proper intersections between the thematic layers.
If you want to produce image objects based exclusively on thematic layer information, you can select a chessboard size larger than your image size between the thematic layers.

4. Chessboard Segmentation
The Chessboard Segmentation algorithm splits the pixel domain or an image object domain into square image objects.
A square grid aligned to the image left and top borders of fixed size is applied to all objects in the domain and each object is cut along these gridlines.
Object Size:
Object Size defines the size of the square grid in pixels. Variables are rounded to the nearest integer

5. Chessboard Segmentation

6. Quadtree-Based Segmentation
The Quadtree-Based Segmentation algorithm splits the pixel domain or an image object domain into a quadtree grid formed by square objects.
A quadtree grid consists of squares with sides each having a power of two and aligned to the image left and top borders. It is applied to all objects in the domain and each object is cut along these gridlines.
The quadtree structure is built so that each square has a maximum possible size and also fulfills the homogeneity criteria defined by the mode and scale parameters.
The maximum square object size is 256 x 256, or 65,536 pixels.

8. Algorithm Parameters Mode Scale
Color: The maximal color difference within each square image object is less than the Scale value
Scale
Scale defines the maximum color difference within each selected image layer inside square image objects. It is only used in conjunction with the Color mode.

9. Contrast Split Segmentation
The Contrast Split Segmentation algorithm segments an image or image object into dark and bright regions.
It is based on a threshold that maximizes the contrast between the resulting bright objects (consisting of pixels with pixel values above the threshold) and dark objects (consisting of pixels with pixel values below the threshold).
The algorithm evaluates the optimal threshold separately for each image object in the image object domain. If the pixel level is selected in the image object domain, the algorithm first executes a chessboard segmentation, then performs the split on each square.

10. It achieves the optimization by considering different pixel values as potential thresholds.
The test thresholds range from the minimum threshold to the maximum threshold, with intermediate values chosen according to the step size and stepping type parameter.
If a test threshold satisfies the minimum dark area and minimum bright area criteria, the contrast between bright and dark objects is evaluated.
The test threshold causing the largest contrast is chosen as the best threshold and used for splitting.

11. Settings Chessboard Tile Size Minimum Threshold Maximum Threshold
This field is available only if pixel level is selected in the Image Object Domain. Enter the chessboard tile size (the default is 1,000).
Minimum Threshold
Enter the minimum gray value to be considered for splitting. The algorithm calculates the threshold for gray values from the minimum threshold value to the maximum threshold value (the default is 0).
Maximum Threshold
Enter the maximum gray value to be considered for splitting. The algorithm calculates the threshold for gray values from the minimum threshold value to the maximum threshold value (the default is 255).
Step Size
Enter the step size by which the threshold increases from the minimum threshold to the maximum threshold. The value is either be added to the threshold or multiplied by the threshold, according to the selection in the Stepping Type field.
The algorithm recalculates a new best threshold each time the threshold is changed by application of the values in the Step Size and Stepping Type fields, until the maximum threshold is reached.
Higher values entered for step size tend to execute more quickly; smaller values tend to achieve a split with a larger contrast between bright and dark objects.

12. Multiresolution Segmentation
The Multiresolution Segmentation algorithm locally minimizes the average heterogeneity of image objects for a given resolution of image objects.
It can be executed on an existing image object level or the pixel level for creating new image objects on a new image object level.
The multiresolution segmentation algorithm consecutively merges pixels or existing image objects.
Thus it is a bottom-up segmentation algorithm based on a pairwise region merging technique.
Multiresolution segmentation is an optimization procedure which, for a given number of image objects, minimizes the average heterogeneity and maximizes their respective homogeneity.

14. The segmentation procedure works according the following rules, representing a mutual- best-fitting approach:
The segmentation procedure starts with single image objects of one pixel and repeatedly merges them in several loops in pairs to larger units as long as an upper threshold of homogeneity is not exceeded locally. This homogeneity criterion is defined as a combination of spectral homogeneity and shape homogeneity. You can influence this calculation by modifying the scale parameter. Higher values for the scale parameter result in larger image objects, smaller values in smaller image objects.
As the first step of the procedure, the seed looks for its best-fitting neighbor for a potential merger.
If best fitting is not mutual, the best candidate image object becomes the new seed image object and finds its best fitting partner.
When best fitting is mutual, image objects are merged.
In each loop, every image object in the image object level will be handled once.
The loops continue until no further merger is possible.

15. Each image object uses the homogeneity criterion to determine the best neighbor to merge with
If the first image object’s best neighbor (red) does not recognize the first image object (gray) as best neighbor, the algorithm moves on (red arrow) with the second image object finding the best neighbor
This branch-to-branch hopping repeats until mutual best fitting partners are found
If the homogeneity of the new image object does not exceed the scale parameter, the two partner image objects are merged.
The procedure continues with another image object’s best neighbor. The procedure iterates until no further image object mergers can be realized without violating the maximum allowed homogeneity of an image object.

16. Composition of Homogeneity Criterion
The object homogeneity to which the scale parameter refers is defined in the Composition of Homogeneity criterion field. In this circumstance, homogeneity is used as a synonym for minimized heterogeneity.
Internally, three criteria are computed: color, smoothness, and compactness.
These three criteria for heterogeneity may be applied in many ways although, in most cases, the color criterion is the most important for creating meaningful objects.
However, a certain degree of shape homogeneity often improves the quality of object extraction because the compactness of spatial objects is associated with the concept of image shape. Therefore, the shape criteria are especially helpful in avoiding highly fractured image object results in strongly textured data (for example radar data).

17. Segmentation Settings
Scale Parameter
The Scale Parameter is an abstract term that determines the maximum allowed heterogeneity for the resulting image objects. For heterogeneous data, the resulting objects for a given scale parameter will be smaller than in more homogeneous data. By modifying the value in the Scale Parameter value you can vary the size of image objects.
Shape
The value of the Shape field modifies the relationship between shape and color criteria; By modifying the Shape criterion,1 you define the color criteria (color = 1 - shape). In effect, by decreasing the value assigned to the Shape field, you define to which percentage the spectral values of the image layers will contribute to the entire homogeneity criterion. This is weighted against the percentage of the shape homogeneity, which is defined in the Shape field.
Changing the weight for the Shape criterion to 1 will result in objects more optimized for spatial homogeneity. However, the shape criterion cannot have a value larger than 0.9, due to the fact that without the spectral information of the image, the resulting objects would not be related to the spectral information at all. The slider bar adjusts the amount of Color and Shape to be used for the segmentation.

18. Compactness
The compactness criterion is used to optimize image objects with regard to compactness. This criterion should be used when different image objects which are rather compact, but are separated from non-compact objects only by a relatively weak spectral contrast. Use the slider bar to adjust the degree of compactness to be used for the segmentation.

20. Spectral Difference Segmentation
The Spectral Difference Segmentation algorithm merges neighboring image objects according to their mean image layer intensity values.
Neighboring image objects are merged if the difference between their layer mean intensities is below the value given by the maximum spectral difference.
This algorithm is designed to refine existing segmentation results, by merging spectrally similar image objects produced by previous segmentations.
It cannot be used to create new image object levels based on the pixel level domain.

21. Segmentation Settings
Maximum Spectral Difference
Define the maximum spectral difference in gray values between image objects that are used during the segmentation. If the difference is below this value, neighboring objects are merged.