This operator can analyze scratch assay images
If desired, it analyzes all images contained in the same directory
One can choose a reference image and all results then are refered to this reference image
Results are the segmented images and a table containing scratch areas as well as area differences
first image file
name of the first (reference) scratch assay image
the only analyzed image if batch mode is deactivated
scratch orientation
horizontally or
vertically
entropy filter size
size of entropy filter mask
increase lets the scratch area decrease
sigma
standard deviation of gauss filter
increase leads to more image smoothing and scratch area tends to decrease
batch mode
should all image files in the given directory be analyzed
maximum iterations
maximum number of iterations for level set segmentation
don't check for scratch presence
don't check for scratch presence prior to segmentation
deactivate, if built-in check for scratch presence fails
alternative: train a new svm model, cf. Scratch Assay SVM Trainer
silent
should results be stored automatically
segmented images and results table are just displayed without saving, if unchecked
use external svm file
should an external svm file be used for classification
the automatic scratch detection uses a built-in support vector machine model to decide whether an image contains a scratch or not, if this detection doesn't work properly an external model file created with the Scratch Assay SVM Trainer can be used for this task
external svm file
absolute path to an external svm model file
Verbose
output some additional information