operator implements the Niblack thresholding algorithm
threshold is based on intensity mean of the image to be processed
details for its calculation can be found in the API of this operator
Input image
the single-channel image to be analyzed
if the image contains multiple channels, only the first one is processed
Mode
processing mode of Niblack operator:
STD: standard NIBLACK algorithm
STD LOCALVARCHECK: standard implementation, but only applied to sliding windows where the local intensity variance exceeds a threshold; windows with lower variance are classified as background
MASKWISE: Niblack is applied mask-wise to the image
WHOLE IMAGE: disables local thresholding, but calculates a single threshold for the whole image
Window size
size of the sliding window
Exclude mask
optional binary mask to exclude image parts from processing
K
parameter K of Niblack algorithm
R
parameter R of Niblack algorithm
activates the enhanced Niblack threshold computation
if set to -1 the standard calculations are done
Variance check neighborhood
size of local neighborhood to decide if Niblack threshold is calculated for a pixel, only used in STD LOCALVARCHECK mode
Variance threshold
intensity variance to be exceeded in local neighborhood for threshold calculation in STD LOCALVARCHECK mode
Verbose
disables/enables output of additional messages on console