FINGERPRINT IMAGE SEGMENTATION USING DATA MANIFOLD CHARACTERISTIC FEATURES

摘要: Automatic fingerprint identification systems (AFIS) have been studied extensively and are widely used for biometric identification. Given its importance, many well-engineered methods developed the different stages that encompass those systems. The first stage of any such system is segmentation actual region from background. This typically achieved by classifying pixels, or blocks based on a set features. In this paper, we describe novel features express underlying manifold topology associated with image patches in local neighborhood. It shown seen high-dimensional space form simple highly regular circular manifold. characterization suggests optimal characterize properties fingerprint. Thus, can be formulated as classification problem deviation expected topology. leads to more robust changes contrast than mean, variance coherence. superior performance proposed eight datasets 2002 2004 Fingerprint Verification Competitions.