Research on Fingerprint Recognition Algorithm Based on Minutiae Matching Pairs

This paper proposes a novel fingerprint minutiae matching structure called “matching pairs”. The matching pairs are constructed based on minutiae that conform to the translation and rotation invariance defined in ISO/IEC 19794-2:2011. The method for establishing matching pairs is simple and computationally efficient. It can partially adapt to the issue of minutiae misalignment caused by fingerprint deformation. By applying a small amount of global constraints, excellent matching results can be achieved. We compared our algorithm with a widely recognized fingerprint matching algorithm on publicly available datasets such as FVC 2000 DB2, as well as our own collected fingerprint dataset. The experimental results demonstrate that our algorithm achieves higher matching accuracy and faster computation speed, providing valuable insights for research and applications in the field of fingerprint recognition.