Adaptive Management of Multimodal Biometrics—A Deep Learning and Metaheuristic Approach
مدیریت تطبیقی بیومتریک چند حالته - یادگیری عمیق و رویکرد فرا مکاشفه ای-2021
This paper introduces the framework for adaptive rank-level biometric fusion: a new approach towards personal authentication. In this work, a novel attempt has been made to identify the optimal design parameters and framework of a multibiometric system, where the chosen biometric traits are subjected to rank-level fusion. Optimal fusion parameters depend upon the security level demanded by a particular biometric application. The proposed framework makes use of a metaheuristic approach towards adaptive fusion in the pursuit of achieving optimal fusion results at varying levels of security. Rank-level fusion rules have been employed to provide optimum performance by making use of Ant Colony Optimization technique. The novelty of the reported work also lies in the fact that the proposed design engages three biometric traits simultaneously for the first time in the domain of adaptive fusion, so as to test the efficacy of the system in selecting the optimal set of biometric traits from a given set. Literature reveals the unique biometric characteristics of the fingernail plate, which have been exploited in this work for the rigorous experimentation conducted. Index, middle and ring fingernail plates have been taken into consideration, and deep learning feature-sets of the three nail plates have been extracted using three customized pre-trained models, AlexNet, ResNet-18 and DenseNet-201. The adaptive multimodal performance of the three nail plates has also been checked using the already existing methods of adaptive fusion designed for addressing fusion at the score-level and decision- level. Exhaustive experiments have been conducted on the MATLAB R2019a platform using the Deep Learning Toolbox. When the cost of false acceptance is 1.9, experimental results obtained from the proposed framework give values of the average of the minimum weighted error rate as low as 0.0115, 0.0097 and 0.0101 for the AlexNet, ResNet-18 and DenseNet-201 based experiments respectively. Results demonstrate that the proposed system is capable of computing the optimal parameters for rank-level fusion for varying security levels, thus contributing towards optimal performance accuracy.© 2021 Elsevier B.V. All rights reserved.
Keywords: Adaptive Biometric Fusion | Ant Colony Optimization | Deep Learning | Fingernail Plate | Multimodal Biometrics | Rank-level Adaptive Fusion
Trustworthy authorization method for security in Industrial Internet of Things
روش مجوز معتبر برای امنیت در اینترنت اشیا صنعتی-2021
Industrial Internet of Things (IIoT) realizes machine-to-machine communication and human–computer inter- action (HCI) through communication network, which makes industrial production automatic and intelligent. Security is critical in IIoT because of the interconnection of intelligent industrial equipment. In IIoT environment, legitimate human–computer interaction can only be performed by authorized professionals, and unauthorized access is not tolerated. In this paper, a reliable authentication method based on biological information is proposed. Specifically, the complete local binary pattern (CLPB) and the statistical local binary pattern (SLPB) are introduced to describe the local vein texture characteristics. Meanwhile, the contrast energy and frequency domain information are regarded as auxiliary information to interpret the finger vein. The distance between the features of the registration image and the test image is used to recognize the finger vein image, so as to realize identity authentication. The experiments are carried out on SDUMLA-FV database and FV-USM database, and results show that the presented method has achieved high recognition accuracy.
Keywords: Industrial Internet of Things (IIoT) | Human–computer interaction (HCI) | Biometric recognition | Comprehensive texture | Security system
Efficient biometric-based identity management on the Blockchain for smart industrial applications
مدیریت هویت مبتنی بر بیومتریک کارآمد در Blockchain برای کاربردهای صنعتی هوشمند-2021
In this work, we propose a new Blockchain-based Identity Management system for smart industry. First, we describe an efficient biometric-based anonymous credential scheme, which supports selective disclosure, suspension/thaw and revocation of credentials/entities. Our system provides non-transferability through a freshly computed hidden biometric attribute, which is generated using a secure fuzzy extractor during each authentication. This mechanism combined with offchain storage guarantees GDPR compliance, which is required for protecting user’s data. We define blinded (Brands) DLRep scheme to provide multi-show unlinkability, which is a lacking feature in Brands’ credential based systems. For larger organizations, we re-design the system by replacing the Merkle Tree with an accumulator to improve scalability. The new system enables auditing by adapting the standard Industrial IoT (IIoT) Identity Management Lifecycle to Blockchain. Finally, we show that the new proposal outperforms BASS, i.e. the most recent blockchain-based anonymous credential scheme designed for smart industry. The computational cost at the user-side (can be a weak IoT device) of our scheme is 8-times less than that of BASS. Thus, our system is more suitable for IIoT.© 2020 Elsevier B.V. All rights reserved.
Keywords: Identity management | Smart industry | Blockchain | Non-transferability | Biometrics | DLRep | Multi-show unlinkability | Selective disclosure | Accumulators
Biometric indices of eleven mangrove fish species from southwest Bangladesh
شاخص های بیومتریک یازده گونه ماهی حرا از جنوب غربی بنگلادش-2021
Biometric indices, i.e. i) length-weight relationships (LWRs), ii) form factor (a3.0), iii) length-frequency distributions (LFDs), and iv) condition factors (relative KR and Fulton’s KF) are considered to be very cru- cial in the assessment of fishery studies as they provide information on fish population growth and coastal habitat well-being. The study of biometric indices of mangrove fish has, however, received little attention. Our research investigates the LFDs, LWRs, a3.0, KR and KF of 395 individuals from nine families (Latidae, Engraulidae, Gobiidae, Mugilidae, Synbranchidae, Schilbeidae, Scatophagidae, Plotosidae, and Terapontidae). The LFDs showed that the lowest total length (TL) was 4.57 cm for Stolephorus tri, and highest TL was 56.20 for Monopterus cuchia. The LWRs showed that the b (allometric coefficient) values ranging from 2.01 (Plotosus canius) to 3.29 (Terapon jarbua), appeared as highly significant (P < 0.001). Moreover, the KR values ranged from 0.80 to 1.36, which indicate a good state of health of the population. Our findings could be useful in updating the FishBase (online database) and tracking mangrove fish spe- cies sustainably.© 2021 National Institute of Oceanography and Fisheries. Hosting by Elsevier B.V. This is an open accessarticle under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Length-weight relationship | Growth | Form factor | Condition | FishBase
Biometric characteristics of Trachinus araneus Cuveir, 1829, Trachinus draco Linnaeus, 1758 and Trachinus radiatus Cuveir, 1829 (Pisces; Trachinidae) from the Egyptian Mediterranean waters
Biometric characteristics of Trachinus araneus Cuveir, 1829, Trachinus draco Linnaeus, 1758 and Trachinus radiatus Cuveir, 1829 (Pisces; Trachinidae) from the Egyptian Mediterranean waters-2021
The biometric characteristics of Trachinus araneus, Trachinus draco and Trachinus radiatus from the Egyptian Mediterranean waters at the West of Alexandria City, were studied. In total, 105, 96 and 55 specimens of these three Fish species were sampled, respectively, by the use of bottom trawls operated in the sectors of El-Dabaa and Sidi Abd El Rahman. The morphological characteristic and related index ratio were determined. The results showed that T. araneus total length varied from 10.9 to 30.0 cm TL with mean length of 19.87 ± 5.43 cm TL and a number of horizontal dots appear to be distributed along the lateral line, T. draco total length was 11.8 to 27.6 cm TL with a mean length of 17.85 ± 4.23 cm TL, and specimens of this Fish species showed yellow vertical oblique lines on the body. On the other hand, T. radiatus is characterized by a total length of 10.6 to 35.0 cm TL with a mean length 18.04 ± 4.87 cm TL, and by circular brown dots spread on the whole body and head. Furthermore, these three Fish species have poisonous spines, one on each operculum, and both spines on the internal border of the dorsal part of each eye orbit. The morphometric regression of each morphometric character showed close agreement between the observed and calculated values.© 2021 National Institute of Oceanography and Fisheries. Hosting by Elsevier B.V. This is an open accessarticle under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Biometric | Characteristics | Trachinus araneus | Trachinus draco | Trachinus radiatus | Mediterranean waters | Egypt
ECG based biometric identification using one-dimensional local difference pattern
شناسایی بیومتریک مبتنی بر ECG با استفاده از الگوی تفاوت محلی تک بعدی-2021
In this work, an enhanced version of 1D local binary pattern is proposed, for the derivation of the most relevant features for ECG-based human recognition. Generally, ECG signal characteristics by nature impose some notable challenges, mostly related to its sensitivity to noises, artifacts, behavioral and emotional disorders and other variability factors. To deal with this critical issue, we use a One-dimensional Local Difference Pattern (1D-LDP) operator to extract the discriminating statistical features from ECG by using the difference between consecutive neighboring samples to capture both the micro and macro patterns information in the heartbeat activity while reducing the local and global variation occurred in ECG over time. To verify its robustness, K-nearest neighbors (KNN) linear support vector machine (SVM) and neural network were performed as the classifier models in this work. Obtained results show that the 1D-LDP operator clearly outperforms existing 1D-LBP variants on MIT-BIH Normal Sinus Rhythm and ECG-ID database.
Keywords: Electrocardiogram | 1D local binary pattern | Biometric | Classification
Human identification driven by deep CNN and transfer learning based on multiview feature representations of ECG
Human identification driven by deep CNN and transfer learning based on multiview feature representations of ECG-2021
Increasingly smart techniques for counterfeiting face and fingerprint traits have increased the potential threats to information security systems, creating a substantial demand for improved security and better privacy and identity protection. The internet of Things (IoT)-driven fingertip electrocardiogram (ECG) acquisition provides broad application prospects for ECG-based identity systems. This study focused on three major impediments to fingertip ECG: the impact of variations in acquisition status, the high computational complexity of traditional convolutional neural network (CNN) models and the feasibility of model migration, and a lack of sufficient fingertip samples. Our main contribution is a novel fingertip ECG identification system that integrates transfer learning and a deep CNN. The proposed system does not require manual feature extraction or suffer from complex model calculations, which improves its speed, and it is effective even when only a small set of training data exists. Using 1200 ECG recordings from 600 individuals, we consider 5 simulated yet potentially practical scenarios. When analyzing the overall training accuracy of the model, its mean accuracy for the 540 chest- collected ECG from PhysioNet exceeded 97.60 %, and for 60 subjects from the CYBHi fingertip-collected ECG, its mean accuracy reached 98.77 %. When simulating a real-world human recognition system on 5 public datasets, the validation accuracy of the proposed model can nearly reach 100 % recognition, outperforming the original GoogLeNet network by a maximum of 3.33 %. To some degree, the developed architecture provides a reference for practical applications of fingertip-collected ECG-based biometric systems and for information network security.
Keywords: Off-the-person | Fingertip ECG biometric | Human identification | Convolutional neural network (CNN) | Transfer learning
Person-identification using familiar-name auditory evoked potentials from frontal EEG electrodes
شناسایی فرد با استفاده از پتانسیل نام-آشنا شنوایی الکترودهای EEG جلو برانگیخته-2021
Electroencephalograph (EEG) based biometric identification has recently gained increased attention of re- searchers. However, state-of-the-art EEG-based biometric identification techniques use large number of EEG electrodes, which poses user inconvenience and consumes longer preparation time for practical applications. This work proposes a novel EEG-based biometric identification technique using auditory evoked potentials (AEPs) acquired from two EEG electrodes. The proposed method employs single-trial familiar-name AEPs extracted from the frontal electrodes Fp1 and F7, which facilitates faster and user-convenient data acquisition. The EEG signals recorded from twenty healthy individuals during four experiment trials are used in this study. Different com- binations of well-known neural network architectures are used for feature extraction and classification. The cascaded combinations of 1D-convolutional neural networks (1D-CNN) with long short-term memory (LSTM) and with gated recurrent unit (GRU) networks gave the person identification accuracies above 99 %. 1D-convolutional, LSTM network achieves the highest person identification accuracy of 99.53 % and a half total error rate (HTER) of 0.24 % using AEP signals from the two frontal electrodes. With the AEP signals from the single electrode Fp1, the same network achieves a person identification accuracy of 96.93 %. The use of familiar-name AEPs from frontal EEG electrodes that facilitates user convenient data acquisition with shorter preparation time is the novelty of this work.
Keywords: Auditory evoked potential | Biometrics | Deep learning | Electroencephalogram | Familiar-name | Person identification
A novel multi-lead ECG personal recognition based on signals functional and structural dependencies using time-frequency representation and evolutionary morphological CNN
تشخیص شخصی نوار قلب ECG مبتنی بر وابستگی های عملکردی و ساختاری سیگنالها با استفاده از نمایش فرکانس زمان و CNN مورفولوژیکی تکاملی-2021
Biometric recognition systems have been employed in many aspects of life such as security technologies, data protection, and remote access. Physiological signals, e.g. electrocardiogram (ECG), can potentially be used in biometric recognition. From a medical standpoint, ECG leads have structural and functional dependencies. In fact, precordial ECG leads view the heart from different axial angles, whereas limb leads view it from various coronal angles. This study aimed to design a personal biometric recognition system based on ECG signals by estimating these latent medical variables. To estimate functional dependencies, within-correlation and cross- correlation in time-frequency domain between ECG leads were calculated and represented in the form of extended adjacency matrices. CNN trees were then introduced through genetic programming for the automated estimation of structural dependencies in extended adjacency matrices. CNN trees perform the deep feature learning process by using structural morphology operators. The proposed system was designed for both closed-set identification and verification. It was then tested on two datasets, i.e. PTB and CYBHi, for performance evaluation. Compared with the state-of-the-art methods, the proposed method outperformed all of them.
Keywords: Biometrics | Electrocardiogram | Functional dependencies | Structural dependencies | Genetic programming | Convolutional neural networks
Evaluation of infrared thermography combined with behavioral biometrics for estrus detection in naturally cycling dairy cows
ارزیابی ترموگرافی مادون قرمز همراه با بیومتریک رفتاری برای تشخیص فحلی در گاوهای شیری دوچرخه سواری طبیعی-2021
Low estrus detection rates (>50%) are associated to extended calving intervals, low economic profit and reduced longevity in Holstein dairy cows. The objective of this study was to evaluate the accuracy of infrared thermography and behavioral biometrics combined as potential estrus alerts in naturally (not induced) cycling dairy cows housed in a tie-stall barn. Eighteen first lactation cows were subjected to transrectal ultrasonography to determine spontaneous ovulation. The dominant follicle (DF) disappearance was used retrospectively as an indirect indicator of ovulation, and to establish the estrus period (48–24 h prior the DF disappearance). Raw skin temperature (Raw IR) and residual skin temperature (Res IR) were recorded using an infrared camera at the Vulva area with the tail (Vtail), Vulva area without the tail (Vnotail), and Vulva’s external lips (Vlips) at AM and PM milking from Day 14 until two days after ovulation was confirmed. Behavioral biometrics were recorded on the same schedule as infrared scan. Behavioral biometrics included large hip movements (L-hip), small hip movements (S-hip), large tail movements and small tail movements to compare behavioral changes between estrus and nonestrus periods. Significant increases in Raw IR skin temperature were observed two days prior to ovulation (Vtail; 35.93 ± 0.27 C, Vnotail; 35.59 ± 0.27 C, and Vlips; 35.35 ± 0.27 C) compared to d 5 (Proestrus; Vtail; 35.29 ± 0.27 C, Vnotail; 34.93 ± 0.31 C, and Vlips; 34.68 ± 0.27 C). No significant changes were found for behavioral parameters with the exception of S-hip movements, which increased at two days before ovulation (d 2; 11.13 ± 1.44 Events/5min) compared to d 5 (7.30 ± 1.02 Events/5min). To evaluate the accuracy of thermal and behavioral biometrics, receiver operating characteristic curve analysis was performed using Youden index (YJ), diagnostic odds ratio, positive likelihood ratio (LR+), Sensitivity, Specificity and Positive predicted value to score the estrus alerts. The greatest accuracy achieved using thermal parameters was for Res IR Vtail PM (YJ = 0.34) and L-hip PM (YJ = 0.27) for behavioral biometrics. Combining thermal and behavioral parameters did not improve the YJ index score but reduced the false-positive occurrence observed by increasing the diagnostic odds ratio (26.62), LR+ (12.47), Specificity (0.97) and positive predicted value (0.90) in a Res IR Vtail PM, S-hip AM, S-hip PM combination. The combination of thermal and behavioral parameters increased the accuracy of estrus detection compared to either thermal or behavioral biometrics, independently in naturally cycling cows during milking.
Keywords: Combined-parameters | First-lactating | Movement-frequency | Preovulation | Skin temperature