# How can I find the distance between different dimensional matrices?

1 view (last 30 days)
wisam kh on 13 Aug 2018
Commented: wisam kh on 14 Aug 2018
Hello
When I extract minutiae of fingerprints, their number is different between one image and another. after preprocessing make thin to ridges
im = bwmorph(im,'thin',Inf);
% now clean up the thin image
im = bwmorph(im,'clean'); % Removes single isolated pixels
im = bwmorph(im,'hbreak'); % Removes H-Breaks
newim = bwmorph(im,'spur'); % Removes spikes
Minutiae is extracted by morphological processes Minutiae is the end points and branches However, due to the difference in the quality of the images and other reasons like different fingerprint size Although the size of the image is fixed, or part of the fingerprint is broken. The features obtained will vary between one image and another, these features are points (x, y and orientation(o)), every fingerprint is represented by a matrix of these features,
How can I find the distance between different dimensional matrices?
thank you
Walter Roberson on 13 Aug 2018
What are the factors that lead to different dimensions? Are you not automatically cropping and resizing what is left to a constant size?
wisam kh on 13 Aug 2018
Well I will add the details

Walter Roberson on 13 Aug 2018
"How can I find the distance between different dimensional matrices?"
You cannot do that; it has no mathematical meaning.
You will need to find a different approach.
##### 3 CommentsShow 1 older commentHide 1 older comment
Walter Roberson on 13 Aug 2018
Fingerprint analysis is not something I have studied.
I would tend to think that one approach would be to use image registration to align the images, and then to extract minutae from both, find locations in common, and compare. I do not know if this approach is practical.
I would suggest that you examine scientific papers on fingerprint analysis, such as at http://scholar.google.com
wisam kh on 14 Aug 2018
Thank you I appreciate your effort