Help regarding optimization tool box for image processing?

Hello I have two images a and b, where b is a rotated and translated image of a. Using Sum of Squared Difference method, I have to find the minimization function using optimization tool box by taking 8 by 8 box and running iterations. How do I go about it?

5 Comments

Write down the function you want to minimize in equation form. Then we can offer advice about how to code it.
Hi! Yeah sure ! Thank you!
cost function = Summation [f(x) - g'(x))^2]
g' is the warped image of f and x is the pixel point. I have to select 8x8 block of pixels, optimize the cost function for n number of iterations.
Can you guide me as to how to code this in optimization toolbox?
So each 8x8 block must have its own distinct warp parameters? What it the form of the warping? Elastic, affine, rigid?
Please post your two images so we don't have to guess, and can give advice faster .
The warping consits of translation and rotation .
The original picture is :
The warped picture is:

Sign in to comment.

 Accepted Answer

Why not just us imregister() which is in the Image Processing Toolbox?

5 Comments

Yeah! But, I must implement it using optimization tool box. Can you please help me with that?
No. I'm sorry but I don't have that toolbox.
Sure, you can frame it as an optimization problem where you minimize the error of the registered images but that's like using a crowbar to tear down a house. It's possible, but it's going to be a lot more work than using a bull dozer.
@Image Analyst Thanks a lot !!!!!!! It works!
@Sean de Wolski Yeah you are right! I am new to Image proc and yet to learn stuffs! Thanks a lot

Sign in to comment.

More Answers (0)

Categories

Asked:

on 9 Oct 2014

Commented:

on 15 Oct 2014

Community Treasure Hunt

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

Start Hunting!