rigid and non rigid image registration using sumulink

using optical flow block Iteratively enables rigid and non rigid image registration
3.5K Downloads
Updated 29 Apr 2009

No License

This optical flow program shows how "optical flow" block from Video and image processing blockset of Simulink is capable dealing with BOTH rigid and non rigid optical flow!

The advantage using this block for optical flow is it's speed and the reason you can easly export it to c (using real time workshop)

To run this demo
------------------
1. run the run_first.m file once
2. run the run_second.mdl file (press play several times)

*press play again and again and again.. every time you hit play the i1 picture will be registered better to i2
*you can change the images in the run_first.m file, using pictures of both rigid and non rigid motion between
*you can change the parameters of the optical flow in the optical flow block in the model run_second.mdl

more info
-----------
1)
please see the stop function of the model to understand the whole process: file->model properties->callback->StopFcn
2)
the reason you have to press play several times (as you want more accuracy) is that: to warp the image, i use function interp2() (within the function ImWarp).
if you know how to implement 2-D data interpolation in simulink (instead uinterp2() function) , please post a modified model.

pictures taken from:
----------------------
1)
non-rigid b-spline grid image registration
http://www.mathworks.com/matlabcentral/fileexchange/20057
2)
High accuracy optical flow
http://www.mathworks.com/matlabcentral/fileexchange/17500

ImWarp function taken from:
-----------------------------
Iterative Pyramidal LK Optical Flow
http://www.mathworks.com/matlabcentral/fileexchange/23142

*By Assaf Laufer
haoman17@gmail.com

Cite As

Assaf laufer (2024). rigid and non rigid image registration using sumulink (https://www.mathworks.com/matlabcentral/fileexchange/23961-rigid-and-non-rigid-image-registration-using-sumulink), MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2008b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
Version Published Release Notes
1.0.0.0