[Exercises]
[Workspace]
Purpose:
The purpose of this lab exercise is to use thresholding
techniques to perform image segmentation.
Activities:
-
Select the file $DIP/data/angio.kdf using the User defined
operator, and display it using the Display Image operator.
-
Glyphs:Input/Output:Data Files:User defined
-
Glyphs:Visualization:Non-Interactive Display:Display Image
-
Use the > (greater than) operator to produce three thresholded
images. Use the following threshold values: 118, 128 and 138. Display
the results.
-
Glyphs:Arithmetic:Comparison Operators:>
-
Glyphs:Visualization:Non-Interactive Display:Display Image
-
Duplicate > two times.
-
Open the > panes and set the "Constant" parameter in each
operator to one of the values given above.
-
Run the operators and display the results.
-
Measure the area of the thresholded images using the Statistics
operator.
-
Glyphs:Data Manip:Analysis & Information:Statistics
-
Glyphs:Input/Output:Information:File Viewer
-
So that only the area is calculated and printed, open the Statistics
operator and select the "positive points" option.
-
You can interactively select the threshold using the Edit Image glyph.
-
Glyphs:Visualization:Interactive Display:Edit Image
-
Click on the "Threshold" option to show the threshold pane
-
Select "Pixel Threshold" in the Threshold window
-
Drag with the mouse the "Lower" horizontal scrooling bar to
dinamically change the threshold level applied to the image.
Exercises
-
Try to segment the image $DIP/data/musc1.kdf into three
components: background, cell type I, and cell type II. Use
thresholding techniques.
-
Overlay the original and thresholded images to show the accuracy
of the segmentation.
Khoros Workspace
Execute the visual program c4s2threshold.wk
Main DIP Menu
DIP Feedback Form
Copyright © 1997-1995 KRI, ISTEC, Ramiro Jordán, Roberto Lotufo. All Rights Reserved