Loren on the Art of MATLAB

March 31st, 2009

Technique for Visual Image Comparison

Recently Steve wrote an interesting post entitled Image Overlay Transparency that I found quite interesting.

Contents

The Idea

The idea is to plot one image on top of the other, exposing pieces of the buried image by making parts of the top image transparent. Here's a teaser for you to look at.

Other Techniques?

Do you have some other techniques for visual image comparison that you find helpful? Post them here.


Get the MATLAB code

Published with MATLAB® 7.8

6 Responses to “Technique for Visual Image Comparison”

  1. Oliver Woodford replied on :

    In my work I often have an image and some 1-d metric for each pixel in the image, which I then want to overlay on the image. I do this by making the image greyscale and using the metric to overlay colour, using a colormap like jet. I use my SC function (on the FEX) to do this for me, using the ‘prob’ or ‘prob_jet’ colormaps.

  2. Oliver Woodford replied on :

    Here is an example of what I was talking about above:
    http://www.robots.ox.ac.uk/~ojw/images/prob_demo.png
    One of the benefits of using SC to do this is that it allows a colorbar to be added (with the correct colours and scale).

  3. Loren replied on :

    Oliver-

    Nice technique!

    –Loren

  4. Mike Garrity replied on :

    This is called a “matrix wipe”. It can be really effective if you add a slider to let the user control the amount of wipe. Also, being able to choose different wipes is useful.

    http://en.wikipedia.org/wiki/Wipe_(transition)

  5. imran replied on :

    Dear all
    I have a question
    If I want to compare two high frequency images, i.e whether they are same or not. Is there any other method other than subtracting the two or taking correlation?

    thanks
    Imran

  6. Ă–mer KAYA replied on :

    Hello all;

    thanks for the information about subtraction;

    I have a duty to complete if i want to detect to filtered images in different program with the same filter and want to detect the differences do i must go with the same extracting method or can i go with something else.

    Becouse if i go with the subtraction there can be minus signs of the process what shall i do with them? or can i do something with them like in code classification (if A(0;0) > 0 then …)

    Thanks all..

    Regards


MathWorks
Loren Shure works on design of the MATLAB language at MathWorks. She writes here about once a week on MATLAB programming and related topics.

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