Figure 1 - uploaded by Luciano Silva
Content may be subject to copyright.
–OCT fingerprint scanned images: (left) A-scan and B-scan; (right) 3D fingerprint image 3D fingerprint construction methods usually employ image acquisition by 2D cameras followed by triangulation 14 or structured-light based images followed by phase detection 15 , both accompanied by 3D reconstruction based on an image model (employing shape from shading, shape from sillouette 16 or a mathematical model as a Lambertian 17 or quadratic surface). In Costa, Bellon and Silva. 18 , a technique to obtain 3D fingerprints from OCT images by detecting the air-skin and dermis-epidermis interfaces has been reported and a database of 3D fingerprints (the first acquired with OCT , to our knowledge) described. These images, 700x700x256 voxels and 7μm resolution (sufficient to scan from newborn babies to adults) 19 were employed in this work to evaluate the proposed matching approaches. Matching techniques can be divided in two groups: a) methods based on 2D fingerprints obtained from 3D reconstructed images and b) 3D Surface-based methods, based on the registration error of 3D surfaces or on geometrical information extracted from 3D images, such as point clouds, curvatures or curve skeletons. In the first group, 2D images of the finger obtained from 3D fingerprints using unrolling techniques are built and 2D matching techniques are used, as proposed by Jain and Abramovich 20 , Chen et al. 21 and Labati et al. 17 .
Source publication
Biometric identification systems have important applications to privacy and security. The most widely used of these, print identification, is based on imaging patterns present in the fingers, hands and feet that are formed by the ridges, valleys and pores of the skin. Most modern print sensors acquire images of the finger when pressed against a sen...
Contexts in source publication
Context 1
... signal intensity (interference signal) is proportional to the reflected light from the internal layers of the sample at the depth of interest and is obtained by processing the information in a computer. The intensity image column (white arrows) in Fig.1 (left) refers to the scanning of a single line of the sample, known as A-scan (axial scan). A transversal cross-section (B-scan) of the image can be obtained by scanning several A-scans in a row. ...
Context 2
... transversal cross-section (B-scan) of the image can be obtained by scanning several A-scans in a row. These can be stacked to build a 3D volume, as shown in Fig.1 (right). In OCT B-scans, the external part of the epidermis (known as stratum corneum), appears on the top as a thin high-intensity layer Fig.1 (left) due to the large refractive index mismatch between the air and the skin. ...
Similar publications
Fingerprinting is an identification method used by enterprises to personalize services for their end-users and detect online fraud or by adversaries to launch targeted attacks. Various tools have been proposed to protect online users from undesired identification probes to enhance the privacy and security of the users. However, we have observed tha...
As biometric applications are gaining popularity, there is increased concern over the loss of privacy and potential misuse of biometric data held in central repositories. We present a biometric authentication framework that constructs a multi-biometric template by layering multiple biomet-rics of a user, such that it is difficult to separate the in...
Citations
... One example is the use of OCT for the purposes of enhancing fingerprint technology. [11][12][13][14] The presence of a subsurface fingerprint (sometimes called the primary fingerprint) at the epidermis-dermis boundary, which is an exact copy of the surface (secondary) fingerprint, means that detection of descrepencies between the two patterns can signify the presence of an artificial fingerprint. Liu et al. has also shown that visualizing the distribution of sweat (eccrine) glands enhances the accuracy of fingerprint detection. ...
Multiple reference optical coherence tomography (MR-OCT) is a new technology ideally suited to low- cost, compact OCT imaging. This modality is an extension of time-domain OCT with the addition of a partial mirror in front of the reference mirror. This enables extended, simultaneous depth scanning with the relatively short scan range of a miniature voice coil motor on which the scanning mirror is mounted. This work details early-stage development of the first iteration of a miniature MR-OCT device. This iteration utilizes a fiber-coupled input from an off-board superluminescent diode (SLD). Dimensions of the module are 40 × 57 mm. Off-the-shelf miniature optical components, voice coil motors and photodetectors are used, with the complexity of design depending on specific applications. The photonic module can be configured as either polarized or non-polarized and can include balanced detection. The results shown in this work are from the non-polarized device. The system was characterized through measurement of the input spectrum, axial resolution and signal-to-noise ratio. Typical B-scans of static and in vivo samples are shown which illustrate the potential applications for such a technology.
... Quantitative research with regards to the effect of force or distortion on fingerprint pattern is in the early stages in forensic science (104, 105). Distortion is also studies in the field of biometrics (106) and its compensation will greatly improve the accuracy of an AFIS system (107). For the fingerprint practitioner, such tools can be of critical importance when it comes to assess whether features claimed to be in correspondence by an expert are truly in line with the expected distortions that we can obtain from marks coming from the same source. ...
The purpose of this paper is to provide an overview of the papers dealing with fingerprints and other body impressions (exception made of bitemarks) that have been published between July 2013 and July 2016. We tried to offer an extensive coverage of the published sources (mainly in English), but remain conscious that exhaustiveness is not possible. The reader will realise that the area is very active and counts with more than 530 publications reviewed for this report. We cover here both matters in relation to the detection of marks (mainly fingermarks) and matters associated with the forensic identification process.
... 3. Differences between OCT fingerprints and conventional surface fingerprints. Results are affected by conventional surface scanning elastic distortion [6]; the lack of a 3D-unwrapping procedure for the OCT fingerprints [29]; differences in ridge-valley thickness; and differences in represented area size and locality; and the effect of degradation on surface fingerprints. ...
The Hybrid fingerprint is a local-quality-specific blend of the surface and internal fingerprints, extracted from optical coherence tomography scans. Owing to its origin, and the manner in which it is obtained, the Hybrid fingerprint is a high-quality and secure fingerprint acquisition solution. This research entails a detailed description of the Hybrid fingerprint, the techniques involved to produce it, and a performance analysis of it. A dataset of 282 fingertip scans was established. Two recognised minutiae extraction and fingerprint matching algorithms were applied to assess the performance of the Hybrid fingerprint. The best equal error rate measured was 1.25%. NIST NFIQ scores and orientation certainty level scores indicated the superiority of the Hybrid fingerprint compared to the internal fingerprint.
... 6 The imaging of subsurface skin layers is accomplished with a technology known as optical coherence tomography (OCT). [7][8][9][10] Proof-of-concept extraction of the internal fingerprint from OCT scans was accomplished by Bossen et al., 8 while da Costa et al. 11 strengthened this hypothesis and also evaluated fingerprint deformation using OCT. However, the presence of speckle noise in OCT scans 12 makes the process of extracting the internal fingerprint challenging and also reduces the quality of the fingerprint itself. ...
Abstract. Optical coherence tomography (OCT) is a high-resolution imaging technology capable of capturing a three-dimensional (3-D) representation of fingertip skin. The papillary junction—a junction layer of skin containing the same topographical features as the surface fingerprint—is contained within this representation. The top edge of the papillary junction contains the topographical information pertinent to the internal fingerprint. Extracting the internal fingerprint from OCT fingertip scans has been shown to be possible. Currently, acquiring the internal fingerprint involves manually defining the region containing it. This manner of definition is inefficient. Perfect knowledge of the location of the papillary junction is hypothesized as achievable. This research details and tests a k-means clustering approach for papillary junction detection. All tested metrics are of a standard comparable to the measured human error. The technique presented in this research is highly successful in detection of the location of the papillary junction. Furthermore, high-quality internal fingerprints are acquired using the coordinates obtained.
... OCT is able to provide a high resolution three-dimensional representation of fingertip skin. This technique has been shown as a potential means of mitigating disadvantages associated with conventional fingerprints [5], [6], [7], [8], [9]. ...
... The root mean squared (RMS)-contrast was used to provide quantitative contrast improvement assessment [19]. The fingerprint extracted from unprocessed OCT images, in a simple averaging fashion (as with previous works [5], [6], [7], [8], [9]), is a low contrast base case. The fingerprint extracted from ground-truth coordinates is a high contrast base case. ...
Current surface fingerprint scanners measure the surface topography of skin, resulting in vulnerabilities to surface skin erosion, distortion due to contact with the scanner, and fingerprint counterfeiting. An improved means of fingerprint acquisition is necessitated in these facts. By employing an imaging technique known as Optical Coherence Tomography to the human fingertip skin, a three-dimensional digital reconstruction of subsurface layers of skin can be used for the extraction of an internal fingerprint. The internal fingerprint is robust towards counterfeiting, damage, and distortion, thus providing a replacement for the surface fingerprint. However, OCT scans are corrupted by speckle noise and have low contrast, resulting in a poor quality fingerprint representation. This research applies image enhancement procedures to OCT scan images to improve internal fingerprint quality. Furthermore, a novel internal fingerprint mapping technique is presented: papillary junction detection followed by defined region mapping. With a RMS-contrast improvement of 97%, this technique yields a much higher quality internal fingerprint when compared to previous techniques.
... OCT is able to provide a high resolution three-dimensional representation of fingertip skin. This technique has been shown as a potential means of mitigating disadvantages associated with conventional fingerprints [5], [6], [7], [8], [9]. ...
... The root mean squared (RMS)-contrast was used to provide quantitative contrast improvement assessment [19]. The fingerprint extracted from unprocessed OCT images, in a simple averaging fashion (as with previous works [5], [6], [7], [8], [9]), is a low contrast base case. The fingerprint extracted from ground-truth coordinates is a high contrast base case. ...
Current surface fingerprint scanners measure the surface topography of skin, resulting in vulnerabilities to surface skin erosion, distortion due to contact with the scanner, and fingerprint counterfeiting. An improved means of fingerprint acquisition is necessitated in these facts. By employing an imaging technique known as Optical Coherence Tomography to the human fingertip skin, a three-dimensional digital reconstruction of subsurface layers of skin can be used for the extraction of an internal fingerprint. The internal fingerprint is robust towards counterfeiting, damage, and distortion, thus providing a replacement for the surface fingerprint. However, OCT scans are corrupted by speckle noise and have low contrast, resulting in a poor quality fingerprint representation. This research applies image enhancement procedures to OCT scan images to improve internal fingerprint quality. Furthermore, a novel internal fingerprint mapping technique is presented: papillary junction detection followed by defined region mapping. With a RMS-contrast improvement of 97%, this technique yields a much higher quality internal fingerprint when compared to previous techniques.
... Without knowing the details of making the artificial fingertips for the fake fingerprints in the dataset, we assume that similar procedures to those reported by Matsumoto et al. [11] were utilized. Therefore, we believe, in accordance with the observations made in [ [32], some extents of elastic deformations would have happened unavoidably during the fabrication processes. The projection from fake 3D fingertips to 2D fingerprint images, as indicated in [ [39] , can also lower matching score or cause nonmatch . ...
Fingerprint is a widely used biometrics. Its
extensive usage motivates imposter to fabricate fake
fingerprints. Vitality detection has been proposed to
prevent counterfeit finger attack. Currently the detection
can be done either during the process of acquiring
fingerprint image or by comparing multiple sequentially
acquired images. It is an ongoing research problem to
detect whether a given fingerprint image is obtained from
a real or a fake fingertip. In this paper we look into the
differences between real and fake fingerprints as the first
step to approach this problem. Specifically, we study the
effects of different imaging sensors on the sizes of
templates and on the matching scores between real and
fake fingerprints. We also compare the fake fingerprints
made from different materials. Experiments are carried
out with two publicly available fingerprint databases and
the findings are reported.
Degraded skin conditions and advanced forgery technologies have reduced the stability and security of identity authentication based on external biometrics of the fingertip, including fingerprints, sweat pores, etc. Biological studies indicate that these external biometrics have their subcutaneous origins. Following the successful application of optical coherence tomography (OCT), the fingertip subcutaneous biometrics such as internal fingerprints and sweat glands have been collected. While researchers have worked on different aspects on subcutaneous biometrics, there is a lack of review of the methods and applications to our best knowledge. A survey on every key step in the subcutaneous biometrics processing is conducted in this paper, including the OCT system design, the extraction of subcutaneous biometrics, the identification and anti-counterfeiting applications of new features. Discussion on the pros and cons of methods, and analysis on the research and application directions, are made in every step. The linkage and difference with the conventional fingerprints are illustrated. Fingertip subcutaneous biometrics will undoubtedly become a new trend in highly secure authentication. This paper intends to provide systematic references and insights of methods and applications and support the researchers for their works on fingertip subcutaneous biometrics.
Optical coherence tomography (OCT), as a non-destructive and high-resolution in vivo imaging technology, has been adapted in fingertip biometric acquisition. It measures on and beneath skin fingertip information as 3D volume data, containing the internal fingerprint, sweat pores and glands, which are robust against poor external skin conditions and resistant to forgery attacks. After introducing the basic principle and system representatives of OCT technology, recent progresses over past decades in the field of fingertip biometrics, especially in biometric identification and deception attack detection, are reviewed and summarized in this paper. Technical limitations and prospects are discussed subsequently. This paper intends to provide systematic references for the design and improvement of OCT system and support the researchers for their works on fingertip biometrics.
