next up previous index
Next: ``Cue-based_WS_relax'' - segment an Up: Segmentation Previous: ``Relaxation_labeling'' Relaxation labeling using

``Cue-based_watershed'' - segment an image using watershed analysis using cues

VFX name:
iCueWatershed
IMPROMPTU Equivalent:
Cue-based_watershed
purpose:
Watershed analysis [] works with the gradient of image and employs a topographical interpretation of gray scale images. High gradient areas, corresponding to region borders, appear as mountain ridges, or watershed lines. Low gradient areas, corresponding to region interiors, appear as flat valleys or catchment basins. Watershed Analysis subdivides the image into catchment basins flanked by watershed lines. In its most basic form, Watershed Analysis tends to severely oversegment the image. The oversegmentaion problem [] can be alleviated by defining either cues or markers. This particular implementation uses interactively defined topological cues to drive a modified form of watershed analysis.
input:
An 8-bit volume (gradient image)

Another two 8-bit volumes, called imagename.b and imagename.c, which contain cue information that has been generated using the INTERSEG package [26].

output:
An 8-bit volume in segmentation volume format (refer to the demask region function, p. [*]) containing the segmented regions or an 8-bit volume containing basins.
parameters:
Default parameters menu is
connectivity = 8/26
voxel_handling = Do_not_modify
min_basin depth = 1
cue_handling = Do_not_use_cues
cue_filename = NO_CUES_SPECIFIED
output_type = Segmentation_volume
basin_filename = NO_BASIN_FN_SPECIFIED

comments:
See below.
1.
This function is much faster than relaxation labeling
2.
See Marker-based_watershed for a comparable function that uses markers to alleviate the oversegmentation problem.


next up previous index
Next: ``Cue-based_WS_relax'' - segment an Up: Segmentation Previous: ``Relaxation_labeling'' Relaxation labeling using
MultiDimensional Image Processing Lab, Penn State University