Adept AdeptSight User Guide - Manual (Page 304)

Segmentation Parameters
There are three types of threshold functions: hard, soft and dynamic.
l
Thresholds are modified in the threshold function Editor, using the mouse.
l The threshold function varies depending on the selected Image Segmentation mode.
Hard Threshold Functions
Hard thresholds produce a blob image in which pixels have only two possible
states: blob or background.
l Background pixels are each attributed a value of 0 in the Blob Image.
l
Blob pixels are each attributed a value of 1 in the Blob Image.
l
Hard thresholding is sometimes referred to as binary thresholding
because all pixels can be considered as having one of two states: blob or
background
Hard thresholding assumes that changes in data values occur at the boundary
between pixels, without allowing a variation in pixels values across blob bound-
aries. Because this is rarely the case, soft thresholding is more often used for
applications.
Hard Thresholding Example
Soft Threshold Functions
Soft thresholds let you use blob detection in cases where boundaries of a blob
region span a few pixels in width, with varying grey-levels between the blob and
the background.
A soft threshold is sloped and covers a range of pixel values that become
weighted pixels once they are processed.
l Weighted pixels are used in the calculation of the blob’s center of mass in
proportion to their weighted value within the soft threshold range.
AdeptSight User's Guide, Version 3.2.x, Updated: 8/23/2012
Page 304