Automated Fhgeqwinf Identification Systems (AFIS) have become a popular tool in many securiq and law enforcement applications. Most of these systems rely on the matching of fingerprints using the position and orientation of ridge endings and bqurcations within the fingerprint image, while this information is suficient for mulching fingerprints in small dafabases, it is not discriminatory enough to provide good resulfs on large collections of fingerprint images. Xkis paper presents a means of obtaining additional discriminately information from fingerprint images by demonstrating a novel method to extract the Iocations of sweat pores from the grayscuie fingerprint image.This is achieved through the implementation of U modified minimum squared error approach. The proposed algorithm is capable of obtaining good results men from images obtained from basic 500 dpi optical live-scan devices despite the common belief that images obtained at this resolution are not of high enough quality. Resulfs of the proposed method are demonstrated on fingerprint images tuken both from a common live-scan device and she inked prints of the NIST 4 databse. An q h n a t i o n of the approach is presented, the results ore discussed, and future research possibilities are put forth.
Some of the current approaches to fingerprint image binarization are investigafed me imporlance of the binariytiun phase of processing as it relates to rhe identIification of accurate minutiae points is established An overview of the most commofii) used approaches in this area is then presented The strengths and weaknesses of these upproaches are explored and suggestions for f u m e reseurch and improvements are set forth. I. INIITRODU CTI ONThe use of biometric information to identify individuals has become a valuable security and forensic tool in recent years [ 13. Of the biometric features available, the fingerprint continues to be one of the most commonly used. This is, in part, due to its technological familiarity, ease of collection, and public acceptance [2], [3]. As a result, the development of faster and more accurate automated fingerprint identification systems (AFIS) continues to be an important research area Zn a typical MIS, two fingerprints are matched by comparing their sets of local features, or minutiae. While these features can be classified into a diverse set of shapes, the two most commonly found are ridge endings and bifurcations. An example of these two local features can be seen in Figure 1.Since ridge endings and bifurcations are the most common features, most AFIS use the locations, angles, and types of these two features when testing for a match. The quality of the MIS is therefore heavily dependent on the system's ability to accurately extract the locations of these points. As a result, most AFIS will pass a fingerprint image through several steps of processing in order to increase the reliability of the information extracted. Figure 2 shows Figure 1. An example of the two most common ridge features: (a) Ridge Endings, and @) Rider Bifurcations a Figure 2. An example of the common processing stages used in minutiae extraction.diagram of some of these stages. Of these processing stages, one of the most critical is binarization. The binarkation stage is responsible for converting the grayscale fmgerpMt image into a black and white image that can be thinned to single pixel width for easy minutiae extraction. If a poor binarization algorithm is used, an accurate skeleton cannot be obtained and the minutiae extraction will not provide an accurate representation of the fingerprint image. On the other hand, if a highly accurate binary image is produced, the skeletonization process will go smoothly and a mini" of invalid minutiae points wilI be introduced, thus maximizing the ability to accumtely match the fingerprint. In the remainder of this paper, the importance of binarization will be explored and some of the common methods that are currently used to perform this critical step in minutiae extraction will be examined. Section 2 will explain the critical nature of the binarization step and look at some common errors that occur. Section 3 will examine some of the current binarization methods, and section 4 will draw some conclusions and set forth some areas for improvement and fi~ture research...
In this paper, we investigate the utility of complex, 2 -0 Gabor jilter for a fingerprint matching system. Previous approaches have used an even symmetric 2-0 Gabor filter to perform fingerprint feature extraction processing. Several processing methods using a complex jilter are explored and compared to the even-symmetric case. Conclusions are drawn based upon both the results of using a complex filter and the added computational costs of doing so.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.