Thin-plate spline (TPS) model has been utilized in fingerprint match to deal with nonlinear deformation by Bazen and
Gerez in 2002. The matching algorithm is composed of two parts, the local match and the global match. In local match,
candidate corresponding points from two minutia sets are obtained by local triangle match which is used for generating
TPS model in global match. However, not all candidate pairs are true corresponding ones in prints, and the spurious pairs
influence the final results. Here we present a novel matching scheme that inserts the graphical model-based confirming
process between the local match and the global match. This middle process not only minimizes the effects of spurious
matched pairs, but also provides the reliable degree of candidate pairs used for similarity. Experiments on databases of
FVC2004 achieve the good performance.
The segmentation of fingerprint images plays an important role in fingerprint recognition. A new algorithm based on
Local Fourier Transform (LFT) for the fingerprint segmentation is proposed in this paper. Firstly, we perform the Local
Fourier Transform on image to get eight independent Local Fourier coefficients per pixel. Then, block features are
extracted by calculating the 2nd, 4th, 6th order moments of the local Fourier coefficients of every pixel in the block. After
that, a Fisher linear discriminant classifier is trained for the classification per block. Finally, mathematical morphology
and region boundary smoothing is applied as postprocessing to obtain compact clusters and to reduce the number of
classification errors. The experimental results on the databases of FVC2004 demonstrate the robustness and the
efficiency of the proposed method.