Fingerprint Alteration

The widespread deployment of Automated Fingerprint Identification Systems (AFIS) in law enforcement and border control applications has heightened the need for ensuring that the security afforded by these systems is not compromised. While several issues related to fingerprint system security have been investigated in the past, including the use of fake fingerprints for masquerading identity, the problem of fingerprint alteration or obfuscation has received no attention in the biometric literature. Fingerprint obfuscation refers to the deliberate alteration of the fingerprint pattern by an individual for the purpose of masking his or her identity. Several cases of fingerprint obfuscation have been described in the media. Existing image quality assessment software cannot detect such altered fingerprints since the implicit image quality during alteration may not change significantly. The goal of this paper is to understand the problem of altered fingerprints and to design solutions that can be used to detect these images. In this regard, this paper makes following contributions: (a) compiling case studies of incidents where individuals were found to have altered their fingerprints for circumventing fingerprint systems; (b) classifying the observed alterations into three broad categories and suggesting possible counter-measures; (c) a method to synthetically generate altered fingerprints in the absence of real-world data; (d) a technique to detect altered fingerprints; and (e) experimental results involving both real-world altered prints and synthetically generated altered prints. Experimental results show the feasibility of the proposed approach in detecting altered fingerprints and highlights the need to further pursue this research agenda.