Latent fingerprint matching using descriptor-based hough transform

Abstract

Abstract-Identifying suspects based on impressions of fingers lifted from crime scenes (latent prints) is a routine procedure that is extremely important to forensics and law enforcement agencies. Latents are partial fingerprints that are usually smudgy, with small area and containing large distortion. Due to these characteristics, latents have a significantly smaller number of minutiae points compared to full (rolled or plain) fingerprints. The small number of minutiae and the noise characteristic of latents make it extremely difficult to automatically match latents to their mated full prints that are stored in law enforcement databases. Although a number of algorithms for matching full to full fingerprints have been published in the literature, they do not perform well on the latent to full matching problem. Further, they often rely on features that are not easy to extract from poor quality latents. In this paper, we propose a new fingerprint matching algorithm which is especially designed for matching latents. The proposed algorithm uses a robust alignment algorithm (descriptor-based Hough transform) to align fingerprints and measures similarity between fingerprints by considering both minutiae and orientation field information. To be consistent with the common practice in latent matching (i.e. only minutiae are marked by latent examiners), the orientation field is reconstructed from minutiae. Since the proposed algorithm relies only on manually marked minutiae, it can be easily used in the law enforcement applications. Experimental results on two different latent databases (NIST SD27 and WVU latent databases) show that the proposed algorithm outperforms two well optimized commercial fingerprint matchers. Further, a fusion of the proposed algorithm and commercial fingerprint matchers leads to improved matching accuracy.

An Efficient Automatic Overlapped Fingerprint Identification and Recognition Using ANFIS Classifier

“…During the classification, we also performed the quality analysis of our databases using NFIQ software, identified the good, bad, and ugly images and applied some preprocessing techniques for further processing. The rank-1 accuracies for various images are evaluated, and the results are compared with Verifinger and Descriptor-Based Hough Transform (DBHT) [8] approaches. The rank-1 accuracies for various qualities are listed in Table 11.…”

“…Most of the existing fingerprint-matching algorithms [1,[3][4][5][8][9][10][11][12][13], and [14] failed to match the overlapped images in an efficient manner. VeriFinger and FingerTec algorithms are unable to split the two fingerprints effectively, and there is no successful extraction of the ridges in the overlapped area.…”

“…Because some of the ridges and valleys are lost during separation, it is difficult to match them with the available template fingerprint images. Hence, we reconstruct the orientation fields using a Hough Transform alignment algorithm [8]. Once the alignment was over, the feature extraction of minutiae and pores (discussion later) was performed, and the minutiae/pore-matching algorithms [20,21] were executed for fingerprint matching.…”

Latent fingerprint indexing: Fusion of level 1 and level 2 features

“…For this purpose, two fingerprint matchers were used (one in-house matcher [17] and one COTS matcher, referred as Matcher 1 and Matcher 2, respectively). The input to the matchers consisted of manually marked minutiae in the latents and automatically extracted minutiae in the rolled prints, which is the same set of minutiae that was used in the indexing experiments.…”

Error correction for micro stereo vision model caused by CCD image sensor rotation

“…Currently research work in this area is rare both at home and abroad now. Abroad, H.Kim [1] and G.Danuser [2][3][4][5][6][7] once carried out the work of three-dimensional shape reconstruction and microscopic positioning by making use of stereo vision which is based on SLM. Domestically, Changchun Institute of Optics [8][9] and Dalian University of Technology [10][11] established micro-operation imaging system with SLM, which is used for micro assembly and micro-manipulation research.…”