Steve on Image Processing

March 7th, 2006

I've received some comments and e-mail asking how to apply the spatial transformation ideas I've been writing about to images. When I started this series, I naively assumed that Image Processing Toolbox users who were interested in spatial transformations would already know about the function imtransform. I was wrong about that, as I've already written.

You can use imtransform and maketform to apply several types of transformations to images, including affine, projective, and custom. In addition, the image registration function cp2tform can make more types of transforms, including linear conformal, polynomial, piecewise linear, and local weighted mean. Since it's going to be a few more weeks before I get into the specific details of imtransform and friends in this blog, I thought I'd provide some pointers to help readers who need the information now. Here are some links:

• Reference page - imtransform
• Reference page - maketform
• Reference page - cp2tform
• Users Guide - Spatial transformations section
• Users Guide - Image registration section
• Demo - Creating a Gallery of Transformed Images
• Demo - Exploring a Conformal Mapping
• Demo - Extracting Slices from a 3-Dimensional MRI Data Set
• Demo - Padding and Shearing an Image Simultaneously
• Demo - Registering an Aeriel Photo to an Orthophoto

09:29 UTC | Posted in Spatial transforms | Permalink | 3 Comments »

You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

3 Responses to “Spatial transformations: Useful toolbox documentation links”

1. Hulya Yalcin replied on March 8th, 2006 at 04:20 UTC :
   

   Hi Steve,

   Are you aware of Peter Kovesi’s web site?
   He has developed some Matlab code to compute
   the dominant motion transformation between
   two successive frames. His code is publicly available
   in his web page.

   It would be nice if you guys integrated those
   matlab routines into matlab as part of the demo
   for spatial transformation.

   To use matlab spatial transformation routines, one
   has to select the control points. But computer vision
   researchers have long been developing algorithms to
   automate this. They first find features on the two
   frames and then they match them. Once they are
   matched, a dominant spatial transformation is
   computed from matched features using RANSAC.

   Hulya

2. Steve replied on March 8th, 2006 at 09:23 UTC :
   

   Hulya – thanks for the tip. We’ll take a look. We wouldn’t use someone else’s code directly, of course – that wouldn’t be ethical or legal.

3. Steve replied on March 31st, 2006 at 08:20 UTC :
   

   Arunkumar – See this post.

These postings are the author's and don't necessarily represent the opinions of The MathWorks.