抽象的な

Latent Fingerprint Matching using Alignment Algorithm Based Hough Transform

G.Rekha, S.V.S Jaya Shyam

Identifying suspects based on latents which are lifted from crime scenes is extremely important to forensics and law enforcement agencies. Latent’s are the partial fingerprints usually smudgy and blurred with small area, and contain large distortion. Due to these characteristics, they have a considerably smaller number of minutiae points compared to full fingerprints and therefore it can be extremely difficult to automatically match latent’s to plain or rolled fingerprints that are stored in law enforcement databases. The goal is to develop a latent matching algorithm that uses only minutiae information. The proposed latent matching algorithm consists of three modules: (i) Align two sets of minutiae with a descriptor based Hough Transform; (ii) Establish the correspondence between minutiae; and (iii) Compute a match score.