Automatic Fingerprint Identification Research Articles

A robust method for partial deformed fingerprints verification using genetic algorithm

Fingerprint verification is a well-researched problem, and automatic fingerprint verification techniques have been successfully adapted to both civilian and forensic applications for many years. However, this technology suffers from the problem of handling incomplete prints and often discards any partial fingerprints obtained. Recent research has begun to delve into the problems of incomplete or partial fingerprints. Genetic algorithm is developed to improvement of deformed ridges and complex distortions in fingerprints verification system. In order to deal with low quality fingerprint images. Experimental results demonstrate the robustness of our algorithm to other methods. And results show that the speed is raised using this method in the overall of the most optimum.

A Robust Fingerprint Indexing Scheme Using Minutia Neighborhood Structure and Low-Order Delaunay Triangles

Fingerprint indexing is a key technique in automatic fingerprint identification systems (AFIS). However, handling fingerprint distortion is still a problem. This paper concentrates on a more accurate fingerprint indexing algorithm that efficiently retrieves the top N possible matching candidates from a huge database. To this end, we design a novel feature based on minutia neighborhood structure (we call this minutia detail and it contains richer minutia information) and a more stable triangulation algorithm (low-order Delaunay triangles, consisting of order 0 and 1 Delaunay triangles), which are both insensitive to fingerprint distortion. The indexing features include minutia detail and attributes of low-order Delaunay triangle (its handedness, angles, maximum edge, and related angles between orientation field and edges). Experiments on databases FVC2002 and FVC2004 show that the proposed algorithm considerably narrows down the search space in fingerprint databases and is stable for various fingerprints. We also compared it with other indexing approaches, and the results show our algorithm has better performance, especially on fingerprints with distortion.

AFIS data compression

This article describes the development and implementation of a data compression algorithm designed specifically for fingerprints, referred to as GBP compression. The algorithm is herein discussed. Data Compression algorithms can be designed for general applications, meaning the input data is unknown. This is more commonly referred to as generic data. [LI01] Or, data compression algorithms can be designed for specific applications. E.g. AFIS [Automated Fingerprint Identification Systems] "When the input is known, higher compression ratios can be achieved with the knowledge of the input data stream." To-date, the highest compression ratio for an unknown input data stream, for all data compression algorithms, is JPEG with an average compression ratio range of 1:17 --- 1:23. [PEN03] The algorithm herein discussed, has a compression ratio range of 1:68 --- 1:92. There is a value, time and place for each design method --- generic or specific --- depending upon a variety of factors. Due to the nature of the use of AFIS for law enforcement and incrimination as well as criminal conviction, there are social issues that make data integrity of paramount concern. This factor influences algorithm selection and design. A lossless algorithm is a must! Also, the nature of AFIS is such that it operates across borders and between states, municipalities and jurisdictions. In addition to the usual issues and resistance to accepting new technology, including software [e.g. resistance to change, fear of system failure, etc.], there are the issues of changing engineering standards [hardware and software] which are governmentally determined as well as governmental policy decisions. Likewise, the portability required in implementing a new algorithm, will have to deal with a variety of hardware and software; as well as be designed to integrate into existing systems. This integration must include the ability to incorporate existing JPEG data files, from existing police databases. This requires a handshaking of standards and conversion programs that maintain data integrity. In addition, there is an in depth discussion of the limits of compression with a novel perspective.

AFIS data compression

This article describes the development and implementation of a data compression algorithm designed specifically for fingerprints, referred to as GBP compression. The algorithm is herein discussed. Data Compression algorithms can be designed for general applications, meaning the input data is unknown. This is more commonly referred to as generic data. [LI01] Or, data compression algorithms can be designed for specific applications. e.g. AFIS [Automated Fingerprint Identification Systems] "When the input is known, higher compression ratios can be achieved with the knowledge of the input data stream." To-date, the highest compression ratio for an unknown input data stream, for all data compression algorithms, is JPEG with an average compression ratio range of 1:17 -- 1:23. [PEN03] The algorithm herein discussed, has a compression ratio range of 1:68 -- 1:92. There is a value, time and place for each design method - generic or specific - depending upon a variety of factors. Due to the nature of the use of AFIS for law enforcement and incrimination as well as criminal conviction, there are social issues that make data integrity of paramount concern. This factor influences algorithm selection and design. A lossless algorithm is a must! Also, the nature of AFIS is such that it operates across borders and between states, municipalities and jurisdictions. In addition to the usual issues and resistance to accepting new technology, including software [e.g. resistance to change, fear of system failure, etc.], there are the issues of changing engineering standards [hardware and software] which are governmentally determined as well as governmental policy decisions. Likewise, the portability required in implementing a new algorithm, will have to deal with a variety of hardware and software; as well as be designed to integrate into existing systems. This integration must include the ability to incorporate existing JPEG data files, from existing police databases. This requires a handshaking of standards and conversion programs that maintain data integrity. In addition, there is an in depth discussion of the limits of compression with a novel perspective.

A New Method of Reproduction of Fingerprints from Corpses in a Bad State of Preservation Using Latex

In view of the problems arising while fingerprinting corpses in a bad state of preservation, especially in case of mummification and carbonization, the authors propose an innovative technique which uses latex film. To better illustrate the potential of the method, two cases where the latex technique was applied successfully are reported: the first one is a mummified body discovered in a shack on the outskirts of Milan and the second one is the case of a burnt corpse found in a car boot. Such a technique is versatile, easy to apply, and allows the operator to work quickly on cadavers without amputating parts, except in rare cases (i.e., burnt bodies with muscle retraction). By the latex technique, a perfect and enduring negative copy of the fingerprint is obtained, ready to be inked and photographed. The numerous copies produced this way can be inked several times allowing for the repeatability of the procedure and this is crucial in cases of problematic legal identification of corpses. In both the cases illustrated, the fingerprints obtained by the latex technique were useful for identification. The quality was good enough for the automatic fingerprint identification system research system to be applied.

Local relative location error descriptor-based fingerprint minutiae matching

Minutiae matching with non-linear distortion is a challenging task in an Automatic Fingerprint Identification System. In this paper, a feature called Local Relative Location Error Descriptor (LRLED) is proposed to overcome non-linear distortion. The LRLED-based algorithm consists of three procedures. Firstly, a pairwise alignment method is proposed to achieve fingerprint alignment. Secondly, a matched minutiae-pair set is obtained with a comparatively loose threshold to reduce false non-matches, which lead not only to most of the corresponding minutiae-pairs, but also to a few non-corresponding minutiae-pairs getting matched. Finally, the LRLED-based similarity measure, which outputs a very high score for a corresponding minutiae-pair but a very low score for a non-corresponding minutiae-pair to reduce false matches, is employed to compute the similarity level between template and test fingerprints. Evaluations on FVC2002 and FVC2004 databases reveal that LRLED is good at distinguishing between corresponding and non-corresponding minutiae-pairs and works well for fingerprint minutiae matching.

Design and implementation of Log-Gabor filter in fingerprint image enhancement

The performance of automatic fingerprint identification system relies heavily on the quality of fingerprint images. Fingerprint enhancement is essential to ensure the robustness of fingerprint identification with respect to the image quality. Gabor filtering is the most popular method in fingerprint enhancement. To overcome the limitations of traditional Gabor filter and promote fingerprint enhancement performance, the Log-Gabor filter is introduced in this paper. The design method and implementation scheme of Log-Gabor filter in fingerprint enhancement are described in detail. The enhancement performance is assessed on standard fingerprint databases. Experimental results show that the proposed Log-Gabor filtering method can effectively improve the fingerprint image quality and promote the reliability of fingerprint identification.

Robust Wireless Transmission of Compressed Latent Fingerprint Images

Maximizing the potential of latent fingerprints from crime scenes in the identification of suspects requires the rapid transfer of the latent from the scene to a remote fingerprint bureau. Transmission over restricted-bandwidth cellular wireless networks requires the latent images to be compressed but without compromising the likelihood of a match being achieved. We present details of experiments to establish the optimum form of compression that provides realistic transmission times and yet does not affect the utility and integrity of the U.K. Fingerprint Service in searching for latent identifications and in archiving unidentified latents on the U.K. national automatic fingerprint identification system (AFIS). Practical aspects of the implemented system, especially in respect to communication and security protocols, are outlined. Finally, we give some details of the operational advantages of this system as it begins to be employed across U.K. police forces.

A Memetic Fingerprint Matching Algorithm

Minutiae point pattern matching is the most common approach for fingerprint verification. Although many minutiae point pattern matching algorithms have been proposed, reliable automatic fingerprint verification remains as a challenging problem, both with respect to recovering the optimal alignment and the construction of an adequate matching function. In this paper, we develop a memetic fingerprint matching algorithm (MFMA) which aims to identify the optimal or near optimal global matching between two minutiae sets. Within the MFMA, we first introduce an efficient matching operation to produce an initial population of local alignment configurations by examining local features of minutiae. Then, we devise a hybrid evolutionary procedure by combining the use of the global search functionality of a genetic algorithm with a local improvement operator to search for the optimal or near optimal global alignment. Finally, we define a reliable matching function for fitness computation. The proposed algorithm was evaluated by means of a series of experiments conducted on the FVC2002 database and compared with previous work. Experimental results confirm that the MFMA is an effective and practical matching algorithm for fingerprint verification. The algorithm is faster and more accurate than a traditional genetic-algorithm-based method. It is also more accurate than a number of other methods implemented for comparison, though our method generally requires more computational time in performing fingerprint matching.

Fingerprint classification using one-vs-all support vector machines dynamically ordered with naı¨ve Bayes classifiers

Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which the SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with naı¨ve Bayes classifiers. This is necessary to break the ties that frequently occur when working with multi-class classification systems that use OVA SVMs. More specifically, it uses representative fingerprint features as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and naı¨ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for five-class classification with the statistical significance. The results show the benefits of integrating different fingerprint features as well as the usefulness of the proposed method in multi-class fingerprint classification.

Identifying the victims of the Indian Ocean tsunami: The role of the private sector

Until the 2004 Indian Ocean tsunami, identifying victims of a mass catastrophe was done largely by police and forensic scientists who tried to match pre-death and post-death data from paper files. The tsunami brought computer databases into the world of forensic identification and led to major involvement from four private-sector companies from Canada, France, Denmark and Norway. Between them, the firms created a system to improve the handling of missing persons’ calls; an automated fingerprint identification system; a system to generate possible matches between pre and post-death data; and a state-of-the-art morgue in Phuket, Thailand. In the past, there has been private-sector involvement in mass death incidents — for example, most funerals are conducted by private firms — but the tsunami marked a major shift to a public-private partnership in an area that has generally been limited to police and forensic scientists.

Pores and Ridges: High-Resolution Fingerprint Matching Using Level 3 Features

Fingerprint friction ridge details are generally described in a hierarchical order at three different levels, namely, Level 1 (pattern), Level 2 (minutia points), and Level 3 (pores and ridge contours). Although latent print examiners frequently take advantage of Level 3 features to assist in identification, Automated Fingerprint Identification Systems (AFIS) currently rely only on Level 1 and Level 2 features. In fact, the Federal Bureau of Investigation's (FBI) standard of fingerprint resolution for AFIS is 500 pixels per inch (ppi), which is inadequate for capturing Level 3 features, such as pores. With the advances in fingerprint sensing technology, many sensors are now equipped with dual resolution (500 ppi/1,000 ppi) scanning capability. However, increasing the scan resolution alone does not necessarily provide any performance improvement in fingerprint matching, unless an extended feature set is utilized. As a result, a systematic study to determine how much performance gain one can achieve by introducing Level 3 features in AFIS is highly desired. We propose a hierarchical matching system that utilizes features at all the three levels extracted from 1,000 ppi fingerprint scans. Level 3 features, including pores and ridge contours, are automatically extracted using Gabor filters and wavelet transform and are locally matched using the Iterative Closest Point (ICP) algorithm. Our experiments show that Level 3 features carry significant discriminatory information. There is a relative reduction of 20 percent in the equal error rate (EER) of the matching system when Level 3 features are employed in combination with Level 1 and 2 features. This significant performance gain is consistently observed across various quality fingerprint images.

Fingerprint minutiae filtering based on multiscale directional information

Automatic identification of humans based on their fingerprints is still one of the most reliable identification methods in criminal and forensic applications, and is widely applied in civil applications as well. Most automatic systems available today use distinctive fingerprint features called minutiae for fingerprint comparison. Conventional feature extraction algorithm can produce a large number of spurious minutiae if fingerprint pattern contains large regions of broken ridges (often called creases). This can drastically reduce the recognition rate in automatic fingerprint identification systems. We can say that for performance of those systems it is more important not to extract spurious (false) minutia even though it means some genuine might be missing as well. In this paper multiscale directional information obtained from orientation field image is used to filter those spurious minutiae, resulting in multiple decrease of their number.

Enhancing security of fingerprints through contextual biometric watermarking

This paper presents a novel digital watermarking technique using face and demographic text data as multiple watermarks for verifying the chain of custody and protecting the integrity of a fingerprint image. The watermarks are embedded in selected texture regions of a fingerprint image using discrete wavelet transform. Experimental results show that modifications in these locations are visually imperceptible and maintain the minutiae details. The integrity of the fingerprint image is verified through the high matching scores obtained from an automatic fingerprint identification system. There is also a high degree of visual correlation between the embedded images, and the extracted images from the watermarked fingerprint. The degree of similarity is computed using pixel-based metrics and human visual system metrics. The results also show that the proposed watermarked fingerprint and the extracted images are resilient to common attacks such as compression, filtering, and noise.

Evidence evaluation in fingerprint comparison and automated fingerprint identification systems—Modelling within finger variability

Recent challenges and errors in fingerprint identification have highlighted the need for assessing the information content of a papillary pattern in a systematic way. In particular, estimation of the statistical uncertainty associated with this type of evidence is more and more called upon. The approach used in the present study is based on the assessment of likelihood ratios (LRs). This evaluative tool weighs the likelihood of evidence given two mutually exclusive hypotheses. The computation of likelihood ratios on a database of marks of known sources (matching the unknown and non-matching the unknown mark) allows an estimation of the evidential contribution of fingerprint evidence. LRs are computed taking advantage of the scores obtained from an automated fingerprint identification system and hence are based exclusively on level II features (minutiae). The AFIS system attributes a score to any comparison (fingerprint to fingerprint, mark to mark and mark to fingerprint), used here as a proximity measure between the respective arrangements of minutiae. The numerator of the LR addresses the within finger variability and is obtained by comparing the same configurations of minutiae coming from the same source. Only comparisons where the same minutiae are visible both on the mark and on the print are therefore taken into account. The denominator of the LR is obtained by cross-comparison with a database of prints originating from non-matching sources. The estimation of the numerator of the LR is much more complex in terms of specific data requirements than the estimation of the denominator of the LR (that requires only a large database of prints from an non-associated population). Hence this paper addresses specific issues associated with the numerator or within finger variability. This study aims at answering the following questions: (1) how a database for modelling within finger variability should be acquired; (2) whether or not the visualisation technique or the choice of different minutiae arrangements may influence that modelling and (3) what is the magnitude of LRs that can be expected from such a model. Results show that within finger variability is affected by the visualisation technique used on the mark, the number of minutiae and the minutiae configuration. They also show that the rates of misleading evidence in the likelihood ratios obtained for one of the configurations examined are low.

Facing the future of biometrics

![][1] For more than a century, criminal investigators have used fingerprints to identify suspects on the basis of evidence left at the crime scene. Whereas fingerprints were once compared visually, today computers speed up the process: the Automated Fingerprint Identification System (AFIS), now used in most Western countries, compares traces found at a crime scene with millions of stored fingerprint images in just a few seconds. Recently, the police in various German cities introduced mobile AFIS, enabling comparisons to be made immediately over mobile communication networks. However, it is not only criminal investigators who use such technology; many commercial access‐control systems use identification systems based on biometrics—the automated recognition of individuals by using behavioural and/or biological characteristics. An increasing demand for safety and security in both the public and private sectors is now driving research in this rapidly growing field. In addition to fingerprints, biometric technologies can make use of a rich variety of anatomical characteristics, such as images of the face, iris or retina, or representations of the geometry of the hand. These systems also use behavioural or physiological characteristics, such as a written signature, voice or the typing pattern on a keyboard. Such functions create an individual signal that can be measured by biometric technology. > …[the] new national ID cards will not only be more secure travel documents—such as the ePass—but will also support new applications that are based on electronic signatures Anyone who has recently travelled to the USA is familiar with the procedure: at US border control, the prints of both index fingers are taken, as is a digital portrait photo. This additional security measure is used not only to combat terrorism but also to monitor residence permits. From the perspective of its operators, it has been a success—after its implementation in 2004, it took only a … [1]: /embed/graphic-1.gif

Orientation feature for fingerprint matching

This paper describes a fingerprint verification algorithm based on the orientation field. The orientation field of a fingerprint image has also been used for image alignment. Area around the core point has been employed as an area of interest for determining the orientation feature map. The algorithm has been tested on two databases (database available from University of Bologna, Biometrics Laboratory and FVC2002). The performance of the algorithm is measured in terms of receiver operating characteristics (ROC). For the University of Bologna database, at ∼ 0 % false acceptance rate (FAR) the genuine acceptance rate (GAR) observed is ∼ 78 % and at ∼ 11 % FAR, GAR is ∼ 97 % . For the FVC2002 database at ∼ 0 % FAR the GAR observed is 75 % and at ∼ 18 % FAR, GAR is 93 % . Proposed algorithm yields better GAR at low FAR with reduced computational complexity. Because of simplicity in computations the algorithm can be easily implemented as an embedded automatic fingerprint identification system (AFIS).

A methodology for identification of fingerprints based on Gabor filter

This work presents an automatic fingerprint identification system using a coefficient map in the power spectrum based on Gabor filter. This filter was created by Dennis Gabor in 1946 and is being used in many biometric technologies. The fingerprint core is found by pixel neighborhood technique. Around the core, the input image is divided in quadrants where information about texture orientation in three different Gabor filter directions is extracted. The angles used are 0, 45 and 90 degrees. It is accomplished convolutions among such Gabor filter orientations with each quadrant of the input image through the Fast Fourier Transform routine. After that, it is needed to identify the image in the frequency domain and the power spectrum is done. The maximum coefficient of the power spectrum is found and a map of forty coefficients is created around this maximum coefficient. That map is summed and divided by the maximum coefficient. Each result is an identifier of the image. Accomplishing such method using three Gabor filter orientations convoluted with the quadrants of the input image results in twelve identifiers. Those twelve identifiers are compared in the matching process. Experimental results demonstrate the good efficiency of the methodology for fingerprint images of good quality.

Review of: Automated Fingerprint Identification Systems (AFIS)

Abstract The proliferation of automated fingerprint identification systems (AFIS) during the past 20 years has arguably been the greatest boon to criminal investigation thus far devised. However, as the author notes during the introduction, very little has been written specifically about these systems other than brief, often overly simplified, references in texts devoted to the more general area of fingerprints. This glaring omission has been thankfully remedied by one of the most comprehensive and enlightening explanations of a fascinating process imaginable, and done so with a marvelously cohesive and highly readable style. One's only regret is that this book had not been written a decade or more ago and offered as a newly revised edition.

Coarse Fingerprint Registration Using Orientation Fields

The majority of traditional research into automated fingerprint identification has focused on algorithms using minutiae-based features. However, shortcomings of this approach are becoming apparent due to the difficulty of extracting minutiae points from noisy or low-quality images. Therefore, there has been increasing interest in algorithms based on nonminutiae features in recent years. One vital stage in most fingerprint verification systems is registration, which involves recovering the transformation parameters that align features from each fingerprint. This paper investigates the use of orientation fields for registration; an approach that has the potential to perform robustly for poor-quality images. Three diverse algorithms have been implemented for the task. The first algorithm is based on the generalized Hough transform, and it works by accumulating evidence for transformations in a discretized parameter space. The second algorithm is based on identifying distinctive local orientations, and using these as landmarks for alignment. The final algorithm follows the path of steepest descent in the parameter space to quickly find solutions that are locally optimal. The performance of these three algorithms is evaluated using an FVC2002 dataset.