|Please use this identifier to cite or link to this publication: http://hdl.handle.net/1926/1352|
This contribution comes with a new filter which highly improve the performance of the contour detection in the binary image, and a second filter which allow the detection of the countours in label images with similar performance.
Gaetan provides classes for a much more efficient method of extracting contours from labeled objects. This is a vast improvement over my previous protocol of using various morphological operations to accomplish this task.Hypothesis:
Gaetan provides a couple of sample images and corresponding results.Open Science:
This contribution conforms in every way to the ideals of open science.Reproducibility:
I downloaded the code. I didn't try to reproduce the results or compile the executables as I had other immediate tasks to which I wanted to apply the filters. I was able to compile the two filters within my own applications, though. I did receive the following compilation warning
itkLabelContourImageFilter.txx:376: warning: unused variable ‘ee1’
which was easily fixed by commenting out that line. My own results were as expected and were achieved much more quickly than with my previous method of using various binary morphological operations.Quality of the data :
N/A as I did not use the data.Interest:
As Gaetan points out, an immediate application is in the signed Maurer distance transform in ITK.
|Categories:||Code speed optimization, Data Representation, Filtering, Image, Language binding, Mathematical Morphology, Programming|
|Keywords:||border detection, binary image, label image, connected components|
Linked Publications more
Reader/Writer for Analyze Object Maps for ITK
by Hawley J., Johnson H.
Diffeomorphic Demons Using ITK's Finite Difference Solver Hierarchy
by Vercauteren T., Pennec X., Perchant A., Ayache N.
Send a message to the author