Though we have many other biometric solutions, fingerprint identification system is widely used for many reasons. Compared to other biometric techniques, the advantages of fingerprint-based identification are as detailed below:
Uniqueness of the fingerprint - the minutiae details of individual ridges and furrows are permanent and unchanging.
The fingerprint is easily captured using low cost fingerprint scanner.
In the case of noisy images, enhancement of curved structures are required. For that purpose we introduce CNN filters that locally adapt their shape to the direction of flow. These CNN filters enable the choice of filter parameters that increase the smoothing power without creating artifacts in the enhanced image.
In this project CNN filters are applied to the curved ridge and valley structures of low quality fingerprint images.