The Nanoscale World

image comparison and subtraction

rated by 0 users
Not Answered This post has 0 verified answers | 3 Replies | 0 Followers

posted on Thu, Jan 7 2010 9:05 PM

Could someone please recommend to me the best (and simplest) way of doing the

(1) Comparing two images (that have drift between them) to find the best
possible overlap (I'm guessing a correlation program of some sort).
(2) Then trimming both images so they show the same area.
(3) Subtracting the resulting images from one another.

I know I can do this "visually" in photoshop etc., but I'm hoping there is a
more precise solution on offer. At present I have the images as both bitmaps,
and also in RHK's native format, *.SM4.

All Replies

replied on Thu, Jan 7 2010 9:06 PM

ou can do the comparison by performing cross-correlation. The coordinates
of the maximum on the correlation image will give you the precise offset,
calculated by statistical method. Then you can subtract one image from
another, using calculated offset. All these operations can be performed in
our software, FemtoScan Online. If you'd send me the original .sm4 files

  • | Post Points: 10
replied on Thu, Jan 7 2010 9:07 PM

WSxM 4.0 Develop 13.0 allows this process. First of all you have to
open both images. Then select in the menu:

(1) Process -> Filter -> Align images. Press on "Select Images..." to
choose the images to align and press OK in both dialog-boxes.
(2) Process -> Zoom or Process -> Multiple Dynamic Zoom to trim the
same area in both images.
(3) Process -> Filter -> Merge. Add both images using -1 as the Image
Factor for one of them.

  • | Post Points: 10
replied on Thu, Jan 7 2010 9:11 PM
You can use the method of counter-scanned images. The method allows correcting drift in SPM 
images (direct and counter) followed by matching the corrected images in a coincidence point
(the common point for both images). Usually, after matching, the corrected images are averaged
within the overlap area but you can subtract one image from another. Both linear and nonlinear drift
correction approaches may be applied, depending on the accuracy required.

See details in the following paper "Automatic drift elimination in probe microscope images based
on techniques of counter-scanning and topography feature recognition" (Measurement Science
and Technology, vol. 18, iss. 3, pp. 907-927, 2007). The paper is available online at

  • | Post Points: 10
Page 1 of 1 (4 items) | RSS
Copyright (c) 2011 Bruker Instruments