How to matching different numbers of features?

Hello

I'm working on a fingerprint matching program. The extracted features are minutiae, each of these features contains the location (x , y) and angle of direction (o). The number of features extracted from images varies, where the features matrix not the same numbers of rows.

How can I match different numbers of features?

If I use the Euclidean distance, the matrix of features must be equal in dimensions, also If I want to train neural network, the matrix of features must be equal in dimensions.

Answer 1

[linda Nankabirwa]

Hello,did you get the solution because am also having the same problem matching features with different dimensions can't figure out how to match.Please help if you got a solution.thanks

[wisam kh]

Sorry I'm still looking for a solution

Answer 2

[Image Analyst]

Try this (a search for point matching algorithm):

https://www.google.com/search?num=20&source=hp&ei=MgjCW9q8MIWy5gL95oywCA&q=point+matching+algorithm&oq=point+matching&gs_l=psy-ab.1.0.0l4j0i22i30l6.1371.3957..7704...0.0..0.112.1307.13j2....2..0....1..gws-wiz.....0..35i39j0i131j0i10.Evs2QK5yS3U

[wisam kh]

Thank you so much for this advice

Answer 3

[linda Nankabirwa]

thanks so much for the reply

[Mahmoud Hassan]

hello, i think that i am working on the same project you are working " minutiae feature extraction " , can i contact with you ?!