Comments on: Dilation, erosion, and the morphological gradient http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/ Steve Eddins manages the Image & Geospatial development team at The MathWorks and coauthored Digital Image Processing Using MATLAB. He writes here about image processing concepts, algorithm implementations, and MATLAB. Fri, 10 Feb 2012 18:55:59 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: Steve http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24413 Steve Wed, 17 Aug 2011 01:59:22 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24413 Tom—Image Processing Toolbox functions do not support spatially varying structuring elements. Tom—Image Processing Toolbox functions do not support spatially varying structuring elements.

]]> By: tom http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24369 tom Thu, 28 Jul 2011 15:15:01 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24369 i have another doubt,is it possible to create a spatially variant structuring element by knowing the orientation of the pixels in an image i have another doubt,is it possible to create a spatially variant structuring element by knowing the orientation of the pixels in an image

]]> By: tom http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24367 tom Thu, 28 Jul 2011 15:05:28 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24367 thank you.. thank you..

]]> By: Steve http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24366 Steve Thu, 28 Jul 2011 14:38:10 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24366 Tom—Sure. imdilate and imerode support arbitrarily shaped, multidimensional structuring elements. See the doc for strel. Tom—Sure. imdilate and imerode support arbitrarily shaped, multidimensional structuring elements. See the doc for strel.

]]> By: tom http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24365 tom Thu, 28 Jul 2011 14:36:33 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24365 can i create a structuring element of my own? can i create a structuring element of my own?

]]> By: Steve http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-24202 Steve Fri, 13 May 2011 14:08:16 +0000 http://blogs.mathworks.com/steve/?p=86#comment-24202 Ricardo—Large structuring elements take longer to compute and can cause detailed structures of interest (as opposed to variations caused by noise) to vanish. Ricardo—Large structuring elements take longer to compute and can cause detailed structures of interest (as opposed to variations caused by noise) to vanish.

]]> By: Ricardo http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-23987 Ricardo Mon, 28 Feb 2011 20:58:15 +0000 http://blogs.mathworks.com/steve/?p=86#comment-23987 Hello Steve. I am currently working on computing the gradient of a synthetic image to which noise has been added. By computing the normal gradient, when I calculate the gaussian curvature of the noisy image, the results are really bad. So I have been trying to compute the morphological gradient instead, that has been giving me better results. However, I've noticed that when computing the directional gradients, by using larger structuring elements, the result is a lot more resistant to noise. My question is, is there a disadvantage to using large structuring elements, and why is it more robust to noise? Thank you. Hello Steve. I am currently working on computing the gradient of a synthetic image to which noise has been added. By computing the normal gradient, when I calculate the gaussian curvature of the noisy image, the results are really bad.

So I have been trying to compute the morphological gradient instead, that has been giving me better results. However, I’ve noticed that when computing the directional gradients, by using larger structuring elements, the result is a lot more resistant to noise.

My question is, is there a disadvantage to using large structuring elements, and why is it more robust to noise? Thank you.

]]> By: Steve http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-23439 Steve Tue, 07 Sep 2010 11:21:54 +0000 http://blogs.mathworks.com/steve/?p=86#comment-23439 Saleem—Yes. I did that a long time ago, and I put them into the Image Processing Toolbox. Saleem—Yes. I did that a long time ago, and I put them into the Image Processing Toolbox.

]]> By: saleem http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-23429 saleem Sun, 05 Sep 2010 10:03:43 +0000 http://blogs.mathworks.com/steve/?p=86#comment-23429 hi , can you write the m-function for erosion and Dilation? hi , can you write the m-function for erosion and Dilation?

]]> By: Steve http://blogs.mathworks.com/steve/2006/09/25/dilation-erosion-and-the-morphological-gradient/#comment-23216 Steve Thu, 24 Jun 2010 15:30:54 +0000 http://blogs.mathworks.com/steve/?p=86#comment-23216 Danish—The Image Processing Toolbox supports morphology on arrays (rectlinear grids) only. You’ll need to develop your own implementations for other geometries. Danish—The Image Processing Toolbox supports morphology on arrays (rectlinear grids) only. You’ll need to develop your own implementations for other geometries.

]]>