This page is a part of, a wiki devoted to computer vision. It focuses on low level computer vision, digital image analysis, and applications. It is designed as an online textbook but the exposition is informal. It geared towards software developers, especially beginners, and CS students. The wiki contains mathematics, algorithms, code examples, source code, compiled software, and some discussion. If you have any questions or suggestions, please contact me directly.

Image registration

From Computer Vision Primer

Jump to: navigation, search

Under construction...

Starting with binary images, always!

A common approach to image registration is minimization of a similarity measure over all possible alignments of the two images. There are a lot of those. However, such a minimization is very time consuming. A direct search for the several best alignments followed by the evaluation of a similarity measure is a more efficient approach.

To align two images you need to establish a correspondence between 2 points in one 2D image and 2 points in the other (3 points if reflection is allowed). The center of mass is normally chosen as the first point. To find the orientation of the image with respect to the center of mass, higher order Moments have been applied. This method has been proven unstable.

Alternatively, one can start with image analysis and align a collection of objects in the first image and a collection in the second image. The centers of mass of these two collections provide the second pair of aligned points. This procedure may be repeated for all collections. An advantage of this approach is that it allows matching images with missing parts. More...

Analysis of microarrays is related...

Other software projects

Personal tools