Systems and methods for point-based image alignment

A method, comprising:

• receiving a first biometric image;
• receiving a second biometric image;
• selecting a first couplet of feature points corresponding to the first biometric image;
• selecting a second couplet of feature points corresponding to the second biometric image;
• calculating a rotation-invariant parameter for the first couplet of feature points and the second couplet of feature points;
• determining that the first couplet of feature points matches the second couplet of feature points based on a difference between the rotation-invariant parameter for the first couplet of feature points and the rotation-invariant parameter for the second couplet of feature points;
• determining a candidate transformation operation that aligns the first couplet of feature points to the second couplet of feature points;
• determining a goodness of fit between a pictorial representation of the first biometric image and a pictorial representation of the second biometric image with the transformation operation applied;
• and
• determining that the candidate transformation aligns the first biometric image to the second biometric image based on the goodness of fit.

An electronic device, comprising:

• a fingerprint sensor configured to capture a first image of a fingerprint;
• a memory storing a second image of a fingerprint;
• and
• a processor configured to perform the steps of:
• receiving a first biometric image;
• receiving a second biometric image;
• selecting a first couplet of feature points corresponding to the first biometric image;
• selecting a second couplet of feature points corresponding to the second biometric image;
• calculating a rotation-invariant parameter for the first couplet of feature points and the second couplet of feature points;
• determining that the first couplet of feature points matches the second couplet of feature points based on a difference between the rotation-invariant parameter for the first couplet of feature points and the rotation-invariant parameter for the second couplet of feature points;
• determining a candidate transformation operation that aligns the first couplet of feature points to the second couplet of feature points;
• determining a goodness of fit between a pictorial representation of the first biometric image and a pictorial representation of the second biometric image with the transformation operation applied;
• and
• determining that the candidate transformation aligns the first biometric image to the second biometric image based on the goodness of fit.