Professional Documents
Culture Documents
4. ECG and fingerprint bimodal authentication
4. ECG and fingerprint bimodal authentication
4. ECG and fingerprint bimodal authentication
A R T I C LE I N FO A B S T R A C T
Keywords: Security research on smart cities mostly focuses on the communication of “things” and rarely focuses on the
Smart cities security interaction between “things” and people. Security breaches at this lower level will render higher-level security
Smart city control access measures ineffective. In this paper, we propose to use a multimodal biometric approach (ECG and Fingerprint) to
Bimodal secure the interaction between “things” and people. ECG biometrics is a relatively new technique and its per-
Biometrics
formance is still inferior to that of fingerprint biometrics. However, as opposed to ECG, fingerprint users touch
Multimodal
objects and inadvertently leave behind their invisible fingerprints marks. Hackers can lift these invisible marks
ECG
Fingerprint and gain illicit access to the devices of their victims. Moreover, ECG based authentication requires users to be
Authentication alive, a quality that can prove important for smart cities. The latter is not true for fingerprint authentication. In
ECG SVM this paper, we combine ECG and Fingerprint features in a bimodal biometric system. Our objective is to enhance
the advantages of both biometric methods while minimizing their weaknesses. For the ECG method, we use an
SVM classifier and for the Fingerprint technique we employ the minutiae extractor and matcher from NBIS. We
fuse the results of the ECG and Fingerprint authentication at the decision level to distinguish between genuine
users and impostors. The literature on ECG and Fingerprint bimodal biometrics is very limited, however, the
obtained results show that this work presents an improvement in terms of EER (Equal Error Rate) compared to
existing work.
⁎
Corresponding author.
E-mail addresses: jarteaga@uottawa.ca (J.S. Arteaga-Falconi), halosman@uottawa.ca (H. Al Osman), elsaddik@uottawa.ca (A. El Saddik).
https://doi.org/10.1016/j.scs.2017.12.023
Received 16 August 2017; Received in revised form 10 December 2017; Accepted 19 December 2017
Available online 03 January 2018
2210-6707/ © 2018 Elsevier Ltd. All rights reserved.
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
275
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
logic control system for matching. They used “Atanassov’s intuitionistic They evaluated the algorithm with two ECG datasets from 25 and 45
fuzzy sets” to determine the number of minutiae that are required for a subjects; each record was 2 min long. They reported an EER of 0.83% to
match. They tested their algorithm with 89 users and attained an ac- 0.86%. Molina, Bruekers, Presura, Damstra, and van der Veen (2007)
curacy of 98.08%. Another study Cao et al. (2012) used the Ant Colony extracted features based on the fiducial points. They calculated a score
Optimization (ACO) scheme to find correspondence between minutiae. for the features in the stored template and input template. To perform
ACO builds a correspondence graph of a promising solution based on authentication, they applied a threshold to the score. For the evalua-
the most prominent features. Ants use this graph to find all minutiae tion, they used ECG records 5 min in length from 10 subjects and ob-
correspondences. They test their algorithm with 800 fingerprints, 100 tained an EER of 2%. Lourenço, Silva, and Fred (2012) applied the k-
fingers with eight images of each finger and report an Equal Error Rate nearest neighbours (k-NN) algorithm and an SVM classifier on ECG
(EER) of 2.79%. A third study Tan and Bhanu (2006) used a genetic features to perform identification and authentication. They used 5-min
algorithm (GA) for matching. They considered the starting population ECG records from 32 subjects. Authentication with the k-NN algorithm
to be the stored template(s) and used the input fingerprint to calculate achieved an EER of 9.39% while authentication with SVM scored a FAR
the fitness. The GA launches an iterative process that produces a new of 0% with a false rejection rate (FRR) of 4.55%. They found that SVM
generation from the previous one through selection, crossover and is better than k-NN for authentication; however, for identification, they
mutation. The GA finishes when it reaches a point where there is no found the k-NN algorithm to be the better method.
change in the fitness values for a number of generations. If the fitness
value is greater than a threshold, there is a match. They tested with 2.2. Multibiometrics
2000 pairs of fingerprints and according to the ROC curve; they obtain a
FAR of 0.05% and a TAR of approximately 84%. A fourth study Kovacs- Biometric systems can use a combination of sensors, algorithms,
Vajna (2000) divided the input fingerprint image into squares and each samples, instances (e.g. left and right iris) or modes (i.e., biometric
square into two triangles. They adjusted the orientation of the triangles traits such as fingerprints and ECG). Such systems are said to implement
in the stored template(s) and the input to minimize distortion. Then, multibiometric methods (Ross, 2009).
they proceeded with a minutiae match. They evaluated the algorithm Multibiometrics has some advantages: it enhances a unibiometric
with 2000 pair of images and reported a TAR of 85% with a FAR of (single biometric trait) system, reduces the probability of the system
0.05%. encountering two users with the same biometric information (non-
For over a decade, the National Institute of Standards and universality characteristic) and can speed up indexing for identification
Technology (NIST) has been working together with the FBI on devel- in a large-scale database. The use of multi trait can narrow down the
oping a biometric software application called the National Background amount of potential matching candidates and focus the search only
Investigation System (NBIS) (Watson et al., 2008). They constantly among a few stored templates. In addition, multibiometrics provides
improve and test the software application on a database of over robustness under noisy environments; if one trait fails, we can use an-
40,000,000 fingerprints. In 2015, they released the latest stable version other trait without disrupting the biometric process (Ross, 2009).
5.0.0, which is open source and publicly available (access is restricted Fusion in a multibiometric process can occur before or after
in some countries). The software has independent modules, including matching (Ross, 2009). Fusion that occurs before the matching stage
the feature extractor and matcher module. The feature extractor can be divided into two types: sensor-level fusion (multi-sensors) and
module uses minutiae features collected using the MINDTCT algorithm feature-level fusion (multi-samples) (Ross, 2009). Sensor-level fusion
(Watson et al., 2008). The matcher module uses the BOZOSRTH3 al- combines several samples of the same biometric trait from different
gorithm (Watson et al., 2008). Maddala, Tangellapally, Bartůněk, and perspectives or take samples of the same biometrics trait with different
Nilsson (2011) evaluated the NBIS with 100 fingerprints and 8 images sensors. Feature-level fusion normalizes, transforms and reduces fea-
of each fingerprint. According to its Detection Error Trade-off (DET) tures from multiple biometric algorithms. Feature-level fusion fits all
graph, they reported an EER of approximately 0.4%. the features into the same vector space to feed a matcher and yield the
result. Fusion that occurs after the matching stage can be divided into
2.1.2. ECG biometrics three categories: match score- = level fusion, rank level fusion and
An ECG displays the electrical activity of the heart. To obtain an decision level fusion. Match score level fusion combines the match
ECG record, a sensor collects, amplifies and filters an electrical signal scores of different biometric to generate a unique match score. Rank
obtained through electrodes in contact with the human body. These Level fusion narrows down the identification possibilities with one
electrodes are placed in specific arrangements called virtual vectors or biometric trait and confirms the identification with the second bio-
leads (Goldberger, 2012). There are 12 lead configurations. Lead-I re- metric trait. Decision Level Fusion uses unibiometrics results (Match or
quires only two electrodes placed on the chest, arms, or hands. ECG is a non-match) to indicate if a final match has been found.
unique signal for every person due to the variation in heart mass, or- Fusion at the decision level does not need the features or scores of
ientation, gender, conductivity, and order of activation of the cardiac the classifiers. This is an advantage when we are combining integrating
muscles (Hoekema, Uijen, & Oosterom, 2001; Van Oosterom, Hoekema, proprietary algorithms into the biometric system. In such scenario, we
& Uijen, 2000). This characteristic makes ECG biometrics possible solely have access to the decision as opposed to the score (Ross, 2009).
(Chan et al., 2008). Literature describes various fusion schemes at the decision level. Ma-
To the best of our knowledge, Biel et al. (2001) presented the first jority vote is the most used one; it returns the decision that the majority
approach for ECG biometrics. They extracted time-, amplitude- and of classifiers choose (Li & Jain, 2009). Linear weighted fusion is an
slope-based features for classification. They concluded that it is possible extension of the majority vote. It assigns a weight to the matchers based
to identify a person using ECG. They also concluded that three elec- on their performances (Li & Jain, 2009). Atrey, Hossain, El Saddik, and
trodes are enough for ECG authentication. In another study, Agrafioti Kankanhalli (2010) point out that weight assignment is the major
and Hatzinakos (2008) used linear discriminant analysis for filtering drawback of linear weighted fusion. Behavior Knowledge Space is a
before matching, then calculated the Euclidean distance and applied a statistical fusing rule that calculates the probability of a decision (match
threshold to determine authenticity. They evaluated the algorithm with or non-match) based on the combination of results from the individual
an ECG database of 27 subjects and a signal length of 2530 s. They matchers. The decision with higher probability in a specific combina-
reported an EER of 0.6%. Chiu, Chuang, and Hsu (2008) applied Haar tion of results is the final decision for the multimodal biometric (Li &
wavelet transformation to the ECG signal and used the obtained coef- Jain, 2009). This technique performs effectively for multi-class (deci-
ficients as ECG features. They calculated the Euclidean distances be- sions and matchers) problems where large datasets are available for
tween the template and the input fingerprint features for matching. training (Raudys & Roli, 2003).
276
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
Previous studies have shown that the ECG trait can be used as 1. Detect the fiducial points (i.e., QRS complex, LP & TP valleys and P
biometric (Agrafioti & Hatzinakos, 2008; Biel et al., 2001; Chiu et al., & T peaks) from each heartbeat in the ECG signal;
2008; Molina et al., 2007). ECG biometric methods do not reach the 2. Extract each complete heartbeat from the signal and align it around
performance level of fingerprint approaches if the ECG signal length is a reference point, (in this case the R peak) (Fig. 3).
kept short. Nevertheless, ECG features are more difficult to spoof. 3. Normalize each ECG heartbeat to eliminate disparities caused by
Related works presented in this section confirm that multibiometric heart rate variations by dividing each feature by the total length of a
systems can perform better than unibiometric ones. Although the lit- heartbeat (Israel et al., 2005);
erature is quite limited on bimodal biometric approaches that combine 4. Extract the ECG features as previously described (Arteaga-Falconi
ECG and fingerprint biometrics, we believe these methods produce et al., 2016) (Fig. 4) to obtain eight features that represent the
promising results. distance in time between peaks and valleys and amplitudes dis-
In this study, we present an ECG and Fingerprint bimodal biometric tances;
authentication algorithm. We fuse the biometric results at the decision 5. Generate the template, a set of all features extracted in step 4, ne-
level and evaluate the system using data obtained from public datasets. cessary for enrollment or authentication;
6. If the system is enrolling a user, the template from step 5 is stored in
3. Design of ECG and fingerprint bimodal algorithm a database or memory. If the system is authenticating a user, the
277
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
278
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
279
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
fragment was 60 s in length and was used for enrollment and the six
by the type of matcher (fingerprint and ECG) that was used first. As
remaining fragments were 4 s in length and were used for authentica-
illustrated in Fig. 6, we first evaluated Fusion Method A and second we
tion. Each fragment corresponded to a random portion of the ECG re-
evaluated Fusion Method B. Because ECG uses SVM, then a threshold is
cord. We did not set a threshold for evaluation because ECG-SVM yields
only applicable to the fingerprint method. We evaluated the bimodal
a matching result (genuine or impostor) and not a score; therefore, we
authentication algorithm for several thresholds (ranging from 0 to 100
only had one operating point and do not generate a DET graph. We
in steps of 1) and we plotted a DET graph (Fig. 9) with the obtained
measure the number of true positives (TP) and false positives (FP) to
results.
calculate the TAR and FAR.
In Fig. 9, we see that fusion method A performs better than fusion
We evaluated the best performing algorithm by measuring the TAR
method B. Fusion method A displayed a FAR of 0.47% with a FRR of
when both algorithms had the same FAR value. The ECG-Threshold
0.46%, which give us an approximate EER value of 0.46%. Method B
matcher (Arteaga-Falconi et al., 2016) reported a TAR of 84.93% with a
displayed a FAR of 6.78% with a FRR of 6.39%, which give us an ap-
FAR of 1.29%. The ECG-SVM matcher only has one operating point,
proximate EER value of 6.58%. The difference between fusion method
therefore the TAR and FAR cannot be calibrated. The ECG – Threshold
A and fusion method B is around 6% in terms of EER. We conclude that
matcher has several operating points (see Fig. 8), therefore we can
fusion method A is the better mechanism for bimodal authentication
adjust the thresholds of this algorithm to generate a DET graph. The
using ECG and fingerprint.
DET graph for the ECG-Threshold matcher shows that a FAR of 7% has
Fusion method A produced better results than fusion method B since
a FRR of 4.5%. This FRR represents a TAR of 95.5%
the fingerprint algorithm uses a threshold on the score, which allows us
(TAR = 100 − FRR). The ECG-SVM achieves a TAR of 100% with a FAR
to maximize the TAR at the expense of the FAR. Then, the ECG algo-
of 7%. Therefore, for a FAR of 7%, the ECG-SVM performs better than
rithm decreases the FAR, which results in a lower EER. This is not the
ECG-Threshold.
case when the ECG matcher executes before that of the fingerprint. The
ECG matcher does not have a threshold to adjust and hence the TAR
4.3. Bimodal algorithm evaluation cannot be maximized by moving the threshold.
The bimodal approach is more effective than the fingerprint or ECG
To evaluate our bimodal biometric algorithm, we used the dataset biometric methods alone. In Section 4.1 we found an EER of 1.18% for
previously described in Section 4.1 for the fingerprint images and the the fingerprint unibiometric scheme. Fusion method A had an EER of
dataset previously described in Section 4.2 for the ECG records. We 0.46%; this shows an improvement of 0.72% for the EER. These results
combine an ECG record with a fingerprint image to generate a “virtual” showed that our bimodal biometric method performs 2.5 times better
subject for the bimodal authentication evaluation. Therefore, our final than the fingerprint unibiometric approach. Moreover, in Section 4.2
database consisted of 73 subjects, where each subject had 7 ECG re- we show that ECG-SVM has a FRR of 0% (FRR = 1 − TAR) with a FAR
cords and 7 fingerprint images. We use one fingerprint image with one of 7%. We cannot generate a DET graph for the ECG-SVM unibiometric
ECG record (60 s long) for enrollment and the other 6 ECG records (4 s scheme because the SVM matcher returns a decision (match or non-
long) with the corresponding 6 fingerprint images for authentication. match) and not a score. Therefore, we examine the FAR of our bimodal
We performed the evaluations with Matlab 2016a running on a authentication method when the FRR is 0%. To do this, we use the DET
Windows 7 64 bits PC with an Intel Core i7 CPU of 2.8 GHz, 8 GB RAM graph of the fusion method A (see Fig. 9) and we obtain a FRR of 0%
memory. We ran our experiments in a batch mode; we enrolled (with when FAR is 2.96%. These results show that our bimodal method also
one 60-s ECG record and one fingerprint image) all the subjects one by performs better than the unibiometric ECG-SVM unibiometric scheme.
one and evaluated them against the non-enrolled subjects. Therefore,
each time we enrolled only one subject we evaluated our algorithm 4.4. Comparison to existing works
against 1 genuine subject and 72 impostors. We repeated the process
with each subject until we enrolled them all. In this section, we compared our proposed scheme to those of
As previously defined in Section 3.2, we evaluated two fusion Manjunathswamy et al. (2015) and Singh and Singh (2012). However,
schemes, fusion Method A and fusion Method B. These schemes differed Manjunathswamy et al. (2015) and Singh and Singh (2012) used
280
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
Table 1
Comparison of Multimodal Results.
Manjunathswamy Singh and Singh This Proposed
et al. (2015) (2012) Work
Table 1 presents the results reported by these works and the results
from our evaluation of the algorithms. We do this to present a fair
comparison. We used the following parameters for all three compared
methods: Enrollment time of 60 s; authentication time of 4 s; number of
Subjects was set to 73, fingerprint data from the FCV2006 database
(Cappelli et al., 2007), and ECG data from the Physionet database. The
physionet database contains several databases. From those we use the
Fig. 10. Multimodal DET Graph: Related Works Comparison. European ST-T Database, the MIT-BIH Normal Sinus Rhythm Data-base,
the MIT-BIH Arrhythmia Database and the QT Database (Goldberger
different databases to evaluate their algorithms. For Manjunathswamy et al., 2000a, 2000b). Our proposed bimodal method has an EER of
et al. (2015), they captured their own ECG and fingerprint dataset. This 0.46%, which is lower than Singh and Singh (2012) (EER of 12.61%)
dataset is not publicly available and the number of testing subjects is and Manjunathswamy et al. (2015) work (EER of 25.25%).
not specified. Singh and Singh (2012) used ECG data from Physionet Furthermore, our approach uses decision level in contrast to the
and fingerprint scores from NIST-BSSR1 (National Institute of Standard approach by Singh and Singh (2012) which fuses with a weight sum
and Technology, 2011). In our work we used ECG data from the same rule at the score level. Fusion at the decision level provides in-
source as Singh and Singh (2012); however, for fingerprint we used the dependence between the matchers (i.e., each matcher works as a uni-
FVC2006 database (Cappelli et al., 2007). biometric until fusion). An independent matcher is the one that gen-
Reported evaluation conditions and datasets are different among all erates a score and makes the matching decision. When fusing at the
three algorithms; therefore, to achieve a proper evaluation, we imple- score level, matchers are not independent because each matcher gen-
ment the other two algorithms and evaluate them with the same da- erates a score and another decision module fuses all these scores and
tasets. makes a matching decision. When fusing at the decision level, matchers
Manjunathswamy et al. (2015) described their results in terms of are independent and another decision module provides a final decision
FRR (0%) and FAR (2.5%). The dashed line in Fig. 10 shows the DET (match or non-match) based on the decision results of the unibiometrics
curve of their algorithm with the dataset we employ to evaluate our matchers. Prabhakar and Jain (2002) found that, in a multibiometric
work. We can observe in Fig. 10 that the EER was approximately 25%. approach, independent matchers perform better; therefore, that is one
Among the many factors a high EER value, one of them is normal- of the reasons that decision-level fusion improves the performance in
ization. Because Manjunathswamy et al. work did not normalize the our multibiometric approach.
ECG signal; changes in the heart rate would have affected the matcher
results. In contrast, our data is composed of users with different heart 5. Conclusion
rates, which affects the response of their algorithm.
The dotted line in Fig. 10 shows the DET curve for the evaluation of Our results shows that fingerprint biometrics perform better than
multimodal algorithm by Singh and Singh (2012). Similarly, to evaluate ECG biometrics; however, if we combine them in a bimodal biometric
this algorithm we used the same datasets used in our algorithm. While method, we significantly improve the results of the system. The EER for
the authors reported an EER of 1.52% for their algorithm, our evalua- the bimodal method was 0.46%., which was is less than the 1.18% EER
tion with our data revealed that their algorithm has an EER closer to of the unibiometric fingerprint. The DET graph of our bimodal bio-
12%. The number of features and length of ECG records are the reason metric showed a FAR of 2.96% with a TAR of 100%. These results are
for the discrepancy between these results. They extracted 20 features better than those of the ECG-SVM unibiometric, where TAR was 100%
from the ECG signal. The extraction of some of these features is not with a FAR of 7%. In terms of security, ECG complements fingerprint
always possible. Filtration of the noise in the ECG signal can render vulnerability of leaving marks behind. With fingerprints, a hacker can
some fiducial points very difficult to locate; therefore, it will prevent to lift marks left behind and gain access to the biometric system. However,
extract a feature. If a feature is missed, then the whole heart beat is the task of hacking the system becomes more difficult with the addition
discarded. This loss of information causes the EER to be higher. Another of ECG. A hacker cannot easily take an ECG record without prior
aspect that affects the EER is the length of the ECG records that they knowledge of the subject. In addition, ECG requires a user to be alive, a
used. The minimum length they used for enrollment and authentication quality that provides a new dimension of security to the fingerprint
is 3 min; some records can be as long as 12 h for enrollment and 12 h for biometrics. This bimodal biometric system is a solution for improving
authentication. In contrast, our dataset consists of 60 s for enrollment the security in the interaction between things and users. This approach
and 4 s for authentication. provides a secure replacement of password management for users in-
Table 1 compares the characteristics of our work with that of the teracting with IoT for smart cities. Users would no longer need to
previously cited works. To compare fairly and accurately the results of concern with memory taxing about their passwords. Smart cities data
all three studies, our evaluation used the same data and parameter sets. collection can be more accurate since there is a reduced risk of im-
postors to use unauthorized IoT devices. While this work focused on
281
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
improving an ECG – Fingerprint bimodal authentication algorithm, the Jain, A. K., Ross, A., & Prabhakar, S. (2004). An introduction to biometric recognition.
topics of security and privacy regarding storage and usage of biometric IEEE Transactions on Circuits and Systems for Video Technology, 14(1), 4–20. http://dx.
doi.org/10.1109/TCSVT.2003.818349.
data are important and should be properly studied in a specialized Jain, A. K., Ross, A., & Pankanti, S. (2006). Biometrics: A tool for information security.
work. For future work, we aim to apply the proposed bimodal algorithm IEEE Transactions on Information Forensics and Security, 1(2), 125–143. http://dx.doi.
to specific applications and further introduce other biometric traits to org/10.1109/TIFS.2006.873653.
Jain, A. K., Ross, A. A., & Nandakumar, K. (2011). Introduction. Introduction to biometrics.
our multibiometric approach. Boston, MA: Springer US1–49. http://dx.doi.org/10.1007/978-0-387-77326-1_1.
Kovacs-Vajna, Z. M. (2000). A fingerprint verification system based on triangular
Acknowledgments matching and dynamic time warping. IEEE Transactions on Pattern Analysis and
Machine Intelligence, 22(11), 1266–1276. http://dx.doi.org/10.1109/34.888711.
Laguna, P., Mark, R. G., Goldberg, A., & Moody, G. B. (1997). A database for evaluation of
We would like to thank the Ecuadorian Government for financially algorithms for measurement of QT and other waveform intervals in the ECG.
support Juan Arteaga-Falconi under the SENESCYT graduate scholar- Computers in cardiology 1997, 673–676.
Multiple classifiers. In S. Z. Li, & A. Jain (Eds.). Encyclopedia of biometricsBoston, MA:
ship program.
Springer US. http://dx.doi.org/10.1007/978-0-387-73003-5_2344 p. 986.
Lourenço, A., Silva, H., & Fred, A. (2012). ECG-based biometrics: A real time classification
References approach. 2012 IEEE international workshop on machine learning for signal processing,
1–6. http://dx.doi.org/10.1109/MLSP.2012.6349735.
Maddala, S., Tangellapally, S. R., Bartůněk, J. S., & Nilsson, M. (2011). Implementation
Agrafioti, F., & Hatzinakos, D. (2008). ECG based recognition using second order statis- and evaluation of NIST Biometric Image Software for fingerprint recognition. ISSNIP
tics. 6th annual communication networks and services research conference (cnsr 2008), biosignals and biorobotics conference 2011, 1–5. http://dx.doi.org/10.1109/BRC.2011.
82–87. http://dx.doi.org/10.1109/CNSR.2008.38. 5740672.
Arteaga-Falconi, J. S., Al Osman, H., & El Saddik, A. (2016). ECG authentication for Maltoni, D. (2015). Fingerprint recognition, overview. In S. Z. Li, & A. K. Jain (Eds.).
mobile devices. IEEE Transactions on Instrumentation and Measurement, 65(3), Encyclopedia of biometrics (pp. 664–668). Boston, MA: Springer US. http://dx.doi.org/
591–600. http://dx.doi.org/10.1109/TIM.;1;2015.2503863. 10.1007/978-1-4899-7488-4_47.
Atrey, P. K., Hossain, M. A., El Saddik, A., & Kankanhalli, M. S. (2010). Multimodal fusion Manjunathswamy, E., Abhishek, A. M., Thriveni, J., Venugopal, R., & Patnaik, L. (2015).
for multimedia analysis: A survey. Multimedia Systems, 16(6), 345–379. http://dx.doi. Multimodal biometric authentication using ECG and fingerprint. International Journal
org/10.1007/s00530-010-0182-0. of Computer Applications, 111(13), 33–39.
Biel, L., Pettersson, O., Philipson, L., & Wide, P. (2001). ECG analysis: A new approach in Markowitz, J. A. (2000). Voice biometrics. Communications of the ACM, 43(9), 66–73.
human identification. Instrumentation and Measurement, IEEE Transactions on, 50(3), http://dx.doi.org/10.1145/348941.348995.
808–812. Martin, A., Doddington, G., Kamm, T., Ordowski, M., & Przybocki, M. (1997). The DET
Bigun, J. (2015). Fingerprint features. In S. Z. Li, & A. K. Jain (Eds.). Encyclopedia of curve in assessment of detection task performance.
biometrics (pp. 609–619). Boston, MA: Springer US. http://dx.doi.org/10.1007/978- Molina, G. G., Bruekers, F., Presura, C., Damstra, M., & van der Veen, M. (2007).
1-4899-7488-4_50. Morphological synthesis of ECG signals for person authentication. 2007 15th
Cao, K., Yang, X., Chen, X., Zang, Y., Liang, J., & Tian, J. (2012). A novel ant colony European signal processing conference, 738–742.
optimization algorithm for large-distorted fingerprint matching. Pattern Recognition, Moody, G. B., & Mark, R. G. (2001). The impact of the MIT-BIH arrhythmia database.
45(1), 151–161. http://dx.doi.org/10.1016/j.patcog.2011.04.016. Engineering in Medicine and Biology Magazine, IEEE, 20(3), 45–50.
Cappelli, R., Ferrara, M., Franco, A., & Maltoni, D. (2007). Fingerprint verification National Institute of Standard and Technology (2011). Biometric score set. Retrieved from
competition 2006. Biometric Technology Today, 15(7–8), 7–9. http://dx.doi.org/10. http://www.itl.nist.gov/iad/894.03/biometricscores/.
1016/S0969-4765(07)70140-6. National Research Council (2003). In S. T. Kent, & L. I. Millett (Eds.). Who goes there?
Chan, A. D. C., Hamdy, M. M., Badre, A., & Badee, V. (2008). Wavelet distance measure Authentication through the lens of privacyWashington, DC: The National Academies
for person identification using electrocardiograms instrumentation and measure- Press. http://dx.doi.org/10.17226/10656.
ment. IEEE Transactions on, 57(2), 248–253. Phillips, P. J., Martin, A., Wilson, C. L., & Przybocki, M. (2000). An introduction evalu-
Chiu, C. C., Chuang, C. M., & Hsu, C. Y. (2008). A novel personal identity verification ating biometric systems. Computer, 33(2), 56–63.
approach using a discrete wavelet transform of the ECG signal. 2008 international Ponemon Institute (2012). Visual privacy productivity study.
conference on multimedia and ubiquitous engineering (mue 2008), 201–206. http://dx. Prabhakar, S., & Jain, A. K. (2002). Decision-level fusion in fingerprint verification.
doi.org/10.1109/MUE.2008.67. Pattern Recognition, 35(4), 861–874. http://dx.doi.org/10.1016/S0031-3203(01)
Clark, G. D., & Lindqvist, J. (2015). Engineering gesture-based authentication systems. 00103-0.
IEEE Pervasive Computing, 14(01), 18–25. Raudys, Š., & Roli, F. (2003). The behavior knowledge space fusion method: Analysis of
Derawi, M. O., Nickel, C., Bours, P., & Busch, C. (2010). Unobtrusive user-authentication generalization error and strategies for performance improvement. In T. Windeatt, &
on mobile phones using biometric gait recognition. Intelligent information hiding and F. Roli (Eds.). Multiple classifier systems: 4th international workshop, MCS 2003
multimedia signal processing (IIH-MSP), 2010 sixth international conference on, Guildford, UK, June 11–13, 2003 proceedings (pp. 55–64). Berlin, Heidelberg: Springer
306–311. Berlin Heidelberg. http://dx.doi.org/10.1007/3-540-44938-8_6.
Espinoza, M., Champod, C., & Margot, P. (2011). Vulnerabilities of fingerprint reader to Ross, A., & Jain, A. K. (2015). Biometrics, overview. In S. Z. Li, & A. K. Jain (Eds.).
fake fingerprints attacks. Forensic Science International, 204(1–3), 41–49. Encyclopedia of biometrics (pp. 289–294). Boston, MA: Springer US. http://dx.doi.org/
Goldberger, A. L., Amaral, L. A. N., Glass, L., Hausdorff, J. M., Ivanov, P. C., Mark, R. G., 10.1007/978-1-4899-7488-4_182.
... Stanley, H. E. (2000a). PhysioBank, PhysioToolkit, and PhysioNet. Circulation, Ross, A. (2009). Multibiometrics. In S. Z. Li, & A. Jain (Eds.). Encyclopedia of biometrics
101(23), e215 LP–e220 LP [Retrieved from http://circ.ahajournals.org/content/101/ (pp. 967–973). Boston, MA: Springer US. http://dx.doi.org/10.1007/978-0-387-
23/e215. abstract]. 73003-5_147.
Goldberger, A. L., Amaral, L. A. N., Glass, L., Hausdorff, J. M., Ivanov, P. C., Mark, R. G., Scotti, F., & Piuri, V. (2010). Adaptive reflection detection and location in iris biometric
... Stanley, H. E. (2000b). PhysioBank, PhysioToolkit, and PhysioNet: Components of images by using computational intelligence techniques Instrumentation and
a new research resource for complex physiologic signals. Circulation, 101(23), Measurement. IEEE Transactions on, 59(7), 1825–1833.
e215–e220. http://dx.doi.org/10.1161/01.CIR.101.23. e215. Singh, Y. N., & Singh, S. K. (2012). Evaluation of electrocardiogram for biometric au-
Goldberger, A. L. (2012). Clinical electrocardiography: A simplified approach (8th ed.). thentication. Journal of Information Security, 3(1), 39–48.
Mosby Elsevier. Singh, Y. N., & Singh, S. K. A. (2013). Identifying individuals using eigenbeat features of
Guillen, E., Alfonso, L., Martinez, K., & Mejia, M. (2012). Vulnerabilities and performance electrocardiogram. Journal of Engineering, 2013(539284), 8.
analysis over fingerprint biometric authentication network. Proceedings of the world Taddei, A., Distante, G., Emdin, M., Pisani, P., Moody, G. B., Zeelenberg, C., & Marchesi,
congress on engineering and computer science, Vol. II, 1–6. [Retrieved from] http:// C. (1992). The European ST-T database: Standard for evaluating systems for the
www.iaeng.org/publication/WCECS2012/WCECS2012_pp 908-913.pdf. analysis of ST-T changes in ambulatory electrocardiography. European Heart Journal,
Guo, Z., Karimian, N., Tehranipoor, M. M., & Forte, D. (2016). Hardware security meets 13(9), 1164–1172 [Retrieved from http://eurheartj.oxfordjournals.org/content/13/
biometrics for the age of IoT. 2016 IEEE international symposium on circuits and systems 9/1164. abstract].
(ISCAS), 1318–1321. http://dx.doi.org/10.1109/ISCAS.2016.7527491. Tan, X., & Bhanu, B. (2006). Fingerprint matching by genetic algorithms. Pattern
Ho Cho, D., Park, K. R., Rhee, D. W., Kim, Y., & Yang, J. (2006). Pupil and iris localization Recognition, 39(3), 465–477. http://dx.doi.org/10.1016/j.patcog.2005.09.005.
for iris recognition in mobile phones. Software engineering, artificial intelligence, net- US Department of State (2002). Safety & security of U.S. borders biometrics. Retrieved from
working, and parallel/distributed computing, 2006. SNPD 2006. Seventh ACIS interna- https://travel.state.gov/content/visas/en/general/border-biometrics.html.
tional conference on, 197–201. Van Oosterom, A., Hoekema, R., & Uijen, G. J. (2000). Geometrical factors affecting the
Hoekema, R., Uijen, G. J. H., & Oosterom, A. V. (2001). Geometrical aspects of the in- interindividual variability of the ECG and the VCG. Journal of Electrocardiology, 33,
terindividual variability of multilead ECG recordings. Biomedical Engineering, IEEE 219–227.
Transactions on, 48(5), 551–559. Wübbeler, G., Stavridis, M., Kreiseler, D., Bousseljot, R.-D., & Elster, C. (2007).
Iancu, I., & Constantinescu, N. (2013). Intuitionistic fuzzy system for fingerprints au- Verification of humans using the electrocardiogram. Pattern Recognition Letters,
thentication. Applied Soft Computing, 13(4), 2136–2142. http://dx.doi.org/10.1016/j. 28(10), 1172–1175. Retrieved from http://www.sciencedirect.com/science/article/
asoc.2012.11.001. pii/S0167865507000463.
Israel, S. A., Irvine, J. M., Cheng, A., Wiederhold, M. D., & Wiederhold, B. K. (2005). ECG Wang, S., & Liu, J. (2011). Biometrics on mobile phone. In D. J. Yang (Ed.). Recent ap-
to identify individuals. Pattern Recognition, 38(1), 133–142. Retrieved from http:// plications in biometrics (pp. 3–22). InTech.
www.sciencedirect.com/science/article/pii/S0031320304002419. Watson, C. I., Garris, M. D., Tabassi, E., Wilson, C. L., McCabe, R. M., Janet, S., & Ko, K.
282
J.S. Arteaga-Falconi et al. Sustainable Cities and Society 40 (2018) 274–283
(2008). User’s guide to NIST biometric image software (NBIS). National Institute of Yang, Y., Wu, L., Yin, G., Li, L., & Zhao, H. (2017). A survey on security and privacy issues
Standards and Technology: National Institute of Standards and Technologyhttp://dx. in internet-of-things. IEEE Internet of Things Journal, 4(5), 1250–1258. http://dx.doi.
doi.org/10.1111/j.1468-0297.2006.01074.x. org/10.1109/JIOT.;1;2017.2694844.
Wayman, J. L. (2015). Biometric verification/identification/authentication/recognition: Yoon, S., Feng, J., & Jain, A. K. (2012). Altered fingerprints: Analysis and detection. IEEE
The terminology. In S. Z. Li, & A. K. Jain (Eds.). Encyclopedia of biometrics (pp. 263– Transactions on Pattern Analysis and Machine Intelligence, 34(3), 451–464. http://dx.
268). Boston, MA: Springer US. http://dx.doi.org/10.1007/978-1-4899-7488-4_206. doi.org/10.1109/TPAMI.2011.161.
283