Model Based Separation of Overlapping Latent Fingerprints

Abstract

Abstract-Latent fingerprints lifted from crime scenes often contain overlapping prints, which are difficult to separate and match by state-of-the-art fingerprint matchers. A few methods have been proposed to separate overlapping fingerprints to enable fingerprint matchers to successfully match the component fingerprints. These methods are limited by the accuracy of the estimated orientation field, which is not reliable for poor quality overlapping latent fingerprints. In this paper, we improve the robustness of overlapping fingerprints separation, particularly for low quality images. Our algorithm reconstructs the orientation fields of component prints by modeling fingerprint orientation fields. In order to facilitate this, we utilize the orientation cues of component fingerprints, which are manually marked by fingerprint examiners. This additional markup is acceptable in forensics, where the first priority is to improve the latent matching accuracy. The effectiveness of the proposed method has been evaluated not only on simulated overlapping prints, but also on real overlapped latent fingerprint images. Compared with available methods, the proposed algorithm is more effective in separating poor quality overlapping fingerprints and enhancing the matching accuracy of overlapping fingerprints.

An Efficient Automatic Overlapped Fingerprint Identification and Recognition Using ANFIS Classifier

“…In our work, a backpropagation algorithm is used as a learning algorithm to train and test the network. In this system, we determined the TAR and ET values for the overlapped and non-overlapped images for various existing algorithms [4][5][6]. The achieved results were compared with the existing approaches as shown in Table 9, which proves that our proposed algorithm has achieved more accuracy.…”

“…In addition, an overlapped image can be split for identification using the existing algorithms [4][5][6] and [17]. An overlapped fingerprint image is partitioned into nonoverlapping blocks with 32 9 32 pixels.…”

“…Particularly during criminal investigations, there is a chance that the fingerprint experts misjudge in identifying a person [3]. Determining the distinct features of prints overlapped with either a finger or the background can be dangerous and time consuming [4][5][6]. Due to overlapping fingerprints, recognition may not be possible even if the fingerprint templates for the images are available.…”

A generative model for fingerprint minutiae

“…As in [21], Θ is divided into singular, Θ s , and continuous (or residual), Θ r , components: Θ = Θ s + Θ r . These two components are regularized separately.…”

Enhancing latent fingerprints on banknotes

“…Since automatic latent segmentation is a very challenging problem, these latent enhancement algorithms require manually marked fingerprint region mask, which is inconvenient. Although several interactive latent enhancement algorithms should be able to process latents on banknotes [5,22], they require even more user inputs and the current trend in the latent fingerprint community to make the enhancement algorithm fully automatic [11].…”