Correct the distortion of fingerprint for recognition application

Abstract EN

Distortion of fingerprint is one of the major causes for false non match.

Whilst this problem affects all fingerprints recognition application, it is especially serious in negative recognition applications, such as watchlist and deduplication applications.

In such applications, malicious users may on purpose distort their fingerprint to avoid the identification.

The proposed method is based on using dilation to fill the gaps that occurred in the ridges (lines) of the fingerprint by the distortion to correcting distortion of fingerprint and extracted the local features (end and bifurcation of ridge) for matching.

The proposed method develops approach for rectifying distortion which is based on a single finger-print image.

The results that find when applying the proposed system were satisfying where the percentage ratio about 91%.

Analyzing the suggested system demonstrates that the system shows sufficient performance parameters.