mia-2dstack-cmeans-presegment

Sysnopis:

mia-2dstack-cmeans-presegment {-i io} {-o string} [ options ...]

Description:

This program first evaluates a sparse histogram of an input image series, then runs a c-means classification over the histogram, and then estimates a per image seeds for later segmentation based on class probabilities. This program accepts only images of eight or 16 bit integer pixels.

Options:

File-IO

-i, --in-file=(input, required); io

input image(s) to be filtered. For supported file types see Plugin type: 2dimage/io

-p, --out-probmap=(output); string

Save probability map to this file

-t, --type=png; stringSTRING

output file name type

-o, --out-labels=(output, required); string

output file name base

-f, --out-flow=(output); string

prefix for flow initialization images

Help & Info

-V, --verbose=warning; dict

verbosity of output, print messages of given level and higher priorities. Supported priorities starting at lowest level are:

info:Low level messages
trace:Function call trace
fail:Report test failures
warning:Warnings
error:Report errors
debug:Debug output
message:Normal messages
fatal:Report only fatal errors
--copyright=(); bool

print copyright information

-h, --help=(); bool

print this help

-?, --usage=(); bool

print a short help

--version=(); bool

print the version number and exit

Parameters

-T, --histogram-thresh=5; float in [0, 50]

Percent of the extrem parts of the histogram to be collapsed into the respective last histogram bin.

-C, --classes=kmeans:nc=3; factory

C-means class initializerC-means class initializer. For supported plug-ins see Plugin type: 1d/cmeans

-S, --seed-threshold=0.9; float in (0, 1)

Probability threshold value to consider a pixel as seed pixel.

-F, --flow-prob-thresh=0.5; float in [0, 1)

Class probability threshold to cut the flow to zero for the source/sink flow connectivity creation

-W, --flow-scale=100; float in (0, inf)

Scaling factor to adjust the flow evaluated from the initial c-means segmentation.

Processing

--threads=-1; int

Maxiumum number of threads to use for processing,This number should be lower or equal to the number of logical processor cores in the machine. (-1: automatic estimation).Maxiumum number of threads to use for processing,This number should be lower or equal to the number of logical processor cores in the machine. (-1: automatic estimation).

Example:

Run the program over images imageXXXX.png with the sparse histogram, threshold the lower 30% bins (if available), run cmeans with two classes on the non-zero pixels and then create the seeds mask as seedXXXX.png.

mia-2dstack-cmeans-presegment -i image.png -o seed -t png --histogram-tresh=30 --classes 2

Author(s):

Gert Wollny