Connected Component Segmentation Application

Brief Description

Compute a connected component segmentation.

Tags

Image Analysis, Segmentation

Long Description

Compute a connected component segmentation, which takes mathematical formula as a neighborhood thresholding criteria.

Parameters

• [param] Input Image (-in): The image to segment.

• [param] Output Shape (-out): The segmentation shape.

• [param] Mask expression (-mask): Mask mathematical expression (only if support image is given)

• [param] Connected Component Expression (-expr): Formula used for connected component segmentation

• [param] Minimum Object Size (-minsize): Min object size (area in pixel)

• [param] OBIA Expression (-obia): OBIA mathematical expression

• [param] DEM Directory (-dem): DEM directory (used to reproject output in WGS84 if input image is in sensor model geometry)

Limitations

None

Authors

OTB-Team

See also

Example of use

• Parameters to set value:

• Input Image: ROI_QB_MUL_4.tif

• Output Shape: ConnectedComponentSegmentation.shp

• Mask expression: "((b1>80)*intensity>95)"

• Connected Component Expression: "distance<10"

• Minimum Object Size: 15

• OBIA Expression: "SHAPE_Elongation>8"

• Command line to execute:

otbcli_ConnectedComponentSegmentation -in ROI_QB_MUL_4.tif -out ConnectedComponentSegmentation.shp -mask "((b1>80)*intensity>95)" -expr "distance<10" -minsize 15 -obia "SHAPE_Elongation>8"