International Journal of Biometrics (19 papers in press)
Personal Authentication based on Wrist and Palm Vein Images
by Abderrahmane Herbadji, Noubeil Guermat, Lahcene Ziet, Mohamed Cheniti, Djamel Herbadji
Abstract: One of the newest promising biometrics researched today is the vein pattern recognition. However, little efforts have been invested in this direction. In this paper, two frameworks focused on a palm and wrist vein based multimodal authentication system are proposed. For the first framework, wrist and palm traits of the same hand are fused, whilst four biometric markers are combined in the second framework In addition, two approaches of score level fusion are applied: (i) transformation-based using sum rule, min-max rules and t-norms, (ii) classifier-based via t-norms. The experimental results on publicly available dataset show that the integration of wrist and palm vein images from both left and right hand gives much improved accuracy than the fusion of two traits of one hand.
Keywords: Biometrics; wrist vein; palm vein; score level fusion ; t-norms; classifiers ,local descriptors.
Feature Selection for Face Authentication Systems: Feature Space Reductionism and QPSO
by Kamal ElDahshan, Eman Elsayed, Ashraf Aboshoha, Ebeid Ali
Abstract: In face authentication systems, the feature selection (FS) process is very important because any feature extractor introduces some irrelevant or noisy features. These features can affect in the performance of such systems. In this paper, a new method is proposed to reduce the computations time in the facial feature selection. Quantum Fourier Transforms (QFT), Discrete Wavelet Transform (DWT) and Scale Invariant Feature Transform (SIFT) are employed separately as feature extractors. The proposed algorithm denoted by FSR_QPSO has two phases: Feature Space Reductionism (FSR) and optimal feature selection based on quantum particle swarm optimization (QPSO). FSR reduces the size of the feature matrix by selecting the best vectors (rows) and rejects the worst. Then QPSO is applied to fetch the optimal features set over the reduced space that contains the best vectors only. The proposed algorithm has been test on ORL and FACE94 databases. The experimental results show that the proposed algorithm reduces feature selection time against the case of using complete feature space. In addition, the proposed system accuracy based on the selected features is better than the corresponding system that uses the whole features' space directly.
Keywords: face based authentication; feature selection; Quantum Fourier Transforms (QFT); Discrete Wavelet Transform (DWT); Scale Invariant Feature Transform (SIFT); Feature Space Reductionism (FSR); Quantum Particle Swarm Optimization (QPSO).
Cancellable Biometric System Based on Linear Combination of Trigonometric Functions with Special Application to Forensic Dental Biometrics
by Mahroosh Banday, Ajaz Hussain Mir
Abstract: Security of data and ensuring privacy has become a serious problem in the practical deployment of biometric systems. Biometric templates in its original form are susceptible to attacks and irrevocable if compromised, and thus need to be protected. However, this template protection which prevents the information loss and hacking of stored templates often comes at the cost of reduced recognition performance. To address this problem, a cancellable biometric template protection scheme based on a linear combination of trigonometric functions, their inverse being multivalued, and used independent key to ensure template security and renewability has been proposed. The proposed template protection approach secures the dental templates stored in forensic records and also maintains the security and recognition performance of the identification system simultaneously. This template protection scheme is simpler and can be used to secure the templates of other biometrics modalities as well. Furthermore, the security analysis of this non invertible transformation has been done by making many attempts to revert the transformed templates back to the original domain which proved effectual in maintaining the noninvertibility. The experimental results indicate that this approach is practical with a low error rate of almost 2% which is an insignificant increase when compared to that of the original templates. Moreover, this system has a high Recognition rate of almost 98% and rank 1 identification rate of 93.33% for original as well as transformed dental templates and it is worth to mention that varying the key value doesnt degrade the system performance in case of template renewability. Thus the system not only increases the recognition accuracy but also adds security to the system simultaneously which are the desired parameters for a successfully secured biometric recognition and identification system.
Keywords: Biometric Template Security; Cancellable Biometrics; Forensic Dentistry; Dental Biometry; Non-invertible Transform.
Improving ear recognition robustness against 3D rotation using statistical modelling based on forensic classification
by Takanari Minamidani, Hideyasu Sai, Daishi Watabe
Abstract: Even though ear shape is used in forensic investigations, an ear identification system for assisting forensic experts is not well developed. One of the reasons for this is the three-dimensional (3D) concave shape of the ear; this changes its two-dimensional (2D) appearance when camera angles change. 3D statistical modeling is necessary to compensate for these changes in 2D appearance. In this study, we aim to increase the number of 3D statistical ear models based on a few forensic classification methods of ear shapes. Experimental evaluation shows that morphological classification based on the antihelix can improve the robustness of ear recognition against the change in camera angles.
Keywords: ear recognition robustness; three-dimensional rotation; statistical modelling; forensic classification; ear shapes; ear identification systems; camera angle change; three-dimensional statistical ear models; morphological classification; antihelix; ear feature points; Gabor features; AKAZE features.
A Robust Approach for Palmprint Biometric Recognition
by Ayushi Mishra, Mohd Aamir Khan, Anand Singh Jalal
Abstract: Biometrics system uses an individuals physical or behavioural feature to recognize an individual. An easy-to-capture biometric modality that could work well with a commodity camera is palmprint. It has coarse lines which can be easily detected using a low resolution camera. To achieve superior recognition results, an accurate segmentation of region of interest is very crucial. In this work, a novel palm print ROI extraction algorithm has been presented which extracts a fixed size region from a full hand image. The proposed approach segments the region of interest which is invariant to the angle between the fingers. Firstly, we detect the palm region and segment it from full hand image and mark it as ROI. After the ROI Extraction, the features are extracted by fusing the BSIF and BRISK features. Finally, the classification is performed by Sparse Representation Classifier (SRC). We have validated the proposed approach on dataset which contains various images of hand at different angel between the fingers. The proposed method had successfully resolved the issues of ROI extraction at different angle between the fingers, and experimental results shows that the proposed approach has successfully achieved the accuracy of 90%.
Keywords: ROI; Palmprint; BSIF; BRISK;.
Real-time single-view face detection and face recognition based on aggregate channel feature
by Michael George, Aswathy Sivan, Babita Roslind Jose, Jimson Mathew
Abstract: A single-view face detector and a novel face recognition method based on the aggregate channel feature (ACF) that work at real-time speeds, suitable in a computing resource-constrained setting are presented in this work. The four stage tree-based face detector is trained on a subset of the AFLW dataset. The face detection performance is analysed using the AFW dataset. The face recogniser uses ACF features along with classification algorithms, either SVM or MLP. The face recogniser is trained and tested on the GATech Face dataset. Our face detector displays comparable performance against the state of the art while working at 29.8 fps. The face recogniser achieves a level of performance that is competitive with other state of the art works. The effect of PCA-based dimension reduction of ACF features on face recognition performance is also studied in this work.
Keywords: aggregate channel feature; ACF; support vector machine; SVM; multi-layer perceptron; MLP; face recognition; face detection.
An accurate hand-based multimodal biometric recognition system with optimised sum rule for higher security applications
by Pallavi D. Deshpande, Prachi Mukherji, Anil S. Tavildar
Abstract: This paper presents a multimodal biometric recognition system using palm print, finger geometry and dorsal palm vein modalities. A specific image acquisition system is designed, fabricated and database of 150 users is created. DWT technique for features extraction is used for palm print and dorsal palm vein modalities. Performance analysis for individual modality is done using receiver operating characteristics and accuracies of 98.775%, 98.45% and 97.60% are obtained respectively for PP, FG and DPV modalities. Further the multimodal system is proposed along with a novel basis for optimally choosing the weights. The score level fusion is done using these optimised weights. Testing, validation and benchmarking of the algorithms are done using our own database, as well as the standard database available on the net. The proposed multimodal system gives enhanced accuracy of 99.80% with very low FAR level of 0.0001.
Keywords: multimodal biometric; MMB; palm print; PP; dorsal palm vein; DPV; finger geometry; FG; false acceptance rate; FAR; genuine acceptance rate; GAR; receiver operating characteristics; ROC; weights optimisation.
Study on soft behavioural biometrics to predict consumer's interest level using web access log
by Nobuyuki Nishiuchi, Seima Aoki
Abstract: This paper presents a soft behavioural biometrics to predict the consumer's interest level in a specific product using access log on websites. The experiments are conducted in a way where the subjects are asked to perform a shopping task on some websites. The comparative analysis is carried out between the interest level of one category product taken from the inquiry, and the access log during the purchasing process on websites. The results show that the behavioural patterns of the web searching and some parameters based on the access log are clearly different depending on the interest level. Moreover, based on the experiments' data, an automatic classification of the interest level is tested using support vector machine (SVM).
Keywords: soft biometrics; behavioural biometrics; consumer's interest level; web access log; web analytics; purchasing process; electric commerce site; automatic classification; support vector machine; SVM.
A common convolutional neural network model to classify plain, rolled and latent fingerprints
by Asif Iqbal Khan, M. Arif Wani
Abstract: Fingerprint classification helps in reducing the number of comparisons during the matching stage in automatic fingerprint identification system. In this study, a convolutional neural network model is proposed for classification of plain, rolled and latent fingerprints. We first propose a new convolutional neural network model initialised with random weights and train the model on fingerprint images. Then we fine-tune two pre-trained convolutional neural network models on fingerprint images. Finally, we compare these three models: two pre-trained models and a custom convolutional neural network model initialised with random weights. We show that pre-trained models can achieve the state-of-the-art results on other similar tasks with no or little fine-tuning. We also show that training data size and depth of the network have a serious impact on the overall performance of deep networks. Dropout is used to enhance the performance of deep networks where the labelled training data is not of sufficient size. All the three models trained on NIST DB4 fingerprint and IIIT-D latent fingerprint databases report good accuracy. By only fine-tuning the pre-trained convolutional neural network model, we get the accuracy of 99%, easily out-performing the state-of-the-art.
Keywords: convolutional neural network; CNN; deep learning; fingerprint classification; ConvNet.
Recent trends of ROI segmentation in iris biometrics: a survey
by Ritesh Vyas, Tirupathiraju Kanumuri, Gyanendra Sheoran, Pawan Dubey
Abstract: Segmentation in iris biometrics deals with the localisation of inner and outer boundaries of the iris and isolation of the region of interest (ROI) from the input eye image. The isolated ROI is further used to extract the meaningful features of iris for its effective representation. That is why accuracy of the segmentation module directly affects the overall accuracy in an iris recognition system. In view of this, the present study provides a comprehensive review of state-of-the-art methods on iris segmentation that were reported after 2011. Iris segmentation approaches based on eye images captured in both visible and near infrared illumination have been reviewed in this paper. The state-of-the-art iris segmentation approaches have been categorised into four broad classes, namely: integro-differential operator (IDO)-based approaches, circular Hough transform (CHT)-based approaches, deep learning-based approaches, and miscellaneous approaches. The sole purpose of this survey is to deliver insights on ROI segmentation, which is a prominent step of iris recognition process, and to suggest prospective research directions to the readers.
Keywords: iris biometrics; region of interest; ROI; iris segmentation; accuracy; near infrared; NIR; visible wavelength; VW.
Special Issue on: Intelligent Computing for the Epidemic Challenges of Biometrics
Design of Embedded Image Teaching System Based on ARM Technology
by Wang Can
Abstract: With the continuous development of multimedia and speech teaching laboratory in Colleges and universities, and the application of digital processing technology, digital voice teaching equipment is attracting more and more attention from domestic universities and instrument manufacturers.Based on the core technology of ARM and DSP dual core technology, the theoretical analysis of digital language learning system is carried out, and the student terminal circuit and teaching software system are designed in detail, and the function and technical index of the whole system are tested.The test results show that the digital speech learning system constructed by the student terminal controlled by ARM and DSP binuclear is fully satisfied with the actual language teaching requirements.
Keywords: ARM technology; embedded image; teaching system.
Linearization Control of AC Permanent Magnet Synchronous Motor Servo System Based on Sensor Technology
by Liu Yongqiu
Abstract: AC permanent magnet synchronous motor servo control system is a complex nonlinear, strong coupling and time-varying system. It has strong uncertainty and nonlinearity, and when the system is running, it also will be disturbed to varying degrees, so the conventional control strategy is difficult to meet the control requirements of high accuracy, high speed and high-performance servo system. This paper adopts a direct feedback linearization control strategy based on sensor technology and uses w, i_das the output of the system to realize the decoupling of the system. In addition, the grey prediction is added to overcome the shortcomings of direct feedback linearization that is sensitive to parameters. Adjusting the uncertain factors block by grey prediction to adapt to the direct feedback linearization rule and achieve the desired effect. MATLAB/Simulink is used to complete the simulation of servo control algorithm. The simulation results show that the direct feedback linearization control is better than the conventional PID control, and the direct feedback linearization control algorithm with grey prediction can improve the performance of the permanent magnet synchronous motor servo control system and can meet the basic requirements of the high-performance servo control system.
Keywords: Permanent magnet synchronous motor; Direct feedback linearization; Linearization control; MATLAB simulation; Sensor.
Research on Optimization of Complex Model of Large-Scale Building Structure Dependent on Adaptive Grey Genetic Algorithm
by Shi Xiaohong
Abstract: Genetic algorithm (GA) is a bionics algorithm based on the biological evolution theory that has received extensive attention in the field of computer science and Optimization in recent years. This paper analyzes and integrates the relevant contents of genetic algorithm and its application in the optimal design of large-scale building structures and analyzes and researches briefly several key factors when the genetic algorithm is applied to the optimal design of large-scale building structures, such as mathematical modeling, constraint condition treatment, generation of initial population and selection of control parameters of genetic algorithm. However, because the simple genetic algorithm is only good at global search, and the local search ability is not enough, it will take quite a long time to achieve the real optimal solution. For the shortcomings of simple genetic algorithm, an improved adaptive grey genetic algorithm is proposed in this paper. The example shows that the obtained adaptive genetic algorithm can improve the convergence and calculation speed when the genetic algorithms is applied to structural optimization design.
Keywords: Large-scale buildings; Structural optimization; Adaptive; Grey genetic algorithm.
Research on Evaluation Model of Deep Foundation Pit Supporting structure in Urban Traffic Tunnel
by He Wenbiao
Abstract: In order to improve the building mechanics performance of the foundation pit supporting structure of urban traffic tunnel and guide the engineering design and construction, the evaluation model of the deep foundation pit supporting structure of urban traffic tunnel is proposed based on mechanical anchoring and linear mechanical loading. The mechanical anchoring method is used to load the support structure of deep foundation pit of urban traffic tunnel. The elastic modulus and yield strength of the support structure of deep foundation pit of urban traffic tunnel are taken as the constraint parameters. The dynamic model of supporting structure in deep foundation pit of tunnel is established. The stiffness degradation increment of urban traffic tunnel deep foundation pit support structure is analyzed by SCBRB component. Under the distribution of bilinear elastic residual stress, the deformation increment of urban traffic tunnel deep foundation pit support structure is obtained. The constitutive relation of dynamic node of deep foundation pit supporting structure of urban traffic tunnel under moving load is constructed, and the tangent elastic modulus of supporting structure of deep foundation pit of tunnel is calculated according to equivalent stress-strain relation. Thus, the mechanical evaluation and optimization design of deep foundation pit supporting structure of urban traffic tunnel are realized. The test results show that the mechanical loading performance of the model is good for the design of the deep foundation pit supporting structure of urban traffic tunnel, and the load capacity of the supporting structure is improved, and the engineering design of the foundation pit supporting structure is optimized.
Keywords: urban traffic tunnel; supporting structure; load; mechanics; evaluation model.
Research on Educational informatization Platform Based on Web2.0
by Wang Zhixue
Abstract: Educational informatization is an important part of national informatization strategy. The realization of educational informatization plays a fundamental, global and lasting role in social development. The Ministry of education degree and graduate education research center takes the guarantee and improvement of graduate education quality as the core goal, proposes the idea of sharing and comprehensive analysis of degree data through the establishment of a unified platform, and establishes the management of educational information of degree and postgraduate students, service system planning and design subject. Under this background, this paper deeply studies the current situation of educational informationization development and the relevant technology of informationization platform and proposes the idea of introducing the informationization platform into the construction of higher education informatization platform. The advent of the Web2.0 era has injected new impetus into the development of network education. Web2.0 has the characteristic of personalization, decentralization, openness, interactivity, and sociality. It corresponds to the educational idea advocated by modern educational theory. Through the method of literature analysis, this paper collates and think about the current educational application research under the Web2.0, and finally through specific case analysis and comparison to verify the views and draw conclusions and provide reference for the future research of Web2.0 and educational informatization platform.
Keywords: Internet; Web2.0; Technology; Network education; Education platform;.
Research on Network Intrusion Detection Security Based on Improved Extreme Learning Algorithm and Neural Network algorithm
by Dai Zhenjun
Abstract: In order to improve the ability of network fuzzy intrusion detection, a network intrusion detection method based on improved extreme learning algorithm and neural network algorithm is proposed to improve the security of the network. ARMA and other linear detection methods are used to construct the network intrusion signal model, and the nonlinear time series and chaos analysis methods are used to extract the feature of network intrusion and big data information analysis. The limit learning method is used for active detection of network intrusion, the adaptive learning method is used for iterative analysis of network intrusion detection, and the correlation characteristic decomposition method is used to improve the convergence of network intrusion detection. The fuzzy neural network algorithm is used to classify the network intrusion features to improve the intrusion detection performance. The simulation results show that this method has high accuracy and strong anti-jamming ability, it has good application value in network security.
Keywords: extreme learning; network intrusion; neural network algorithm; detection; nonlinear time series analysis.
RESEARCH ON MANIPULATOR MOTION CONTROL BASED ON NEURAL NETWORK ALGORITHM
by Shi Qiongyan, Zhang Jianghua
Abstract: The manipulator is the new artificial intelligence device, and its motion control is the basis for ensuring the stability of the manipulator's attitude. The traditional manipulator motion control adopts the static neuron control method, which will lead to small disturbance in the attitude control of the manipulator, and cause the stable motion performance of the manipulator. A motion control algorithm for manipulator is proposed based on variable structure fuzzy PID neural network. The coordinate system structure description and manipulator dynamics analysis of the controlled system are carried out. The motion control algorithm of the manipulator is improved by using variable structure PID neural network control and adaptive disturbance suppression method. Combined with the strict feedback control method, the motion error of the manipulator is compensated, and the steady-state error is corrected by the adaptive inertial compensation method to realize the motion control optimization of the manipulator. The simulation results show that the motion control algorithm of the manipulator has better positioning performance and better control stability, reduces the steady-state error and improves the control stability.
Keywords: neural network; manipulator; fuzzy PID; neural network.
Research on Logistics Distribution path Analysis based on artificial Intelligence algorithm
by Yao Cuiping
Abstract: Logistics distribution path optimization model design is the key to ensure the smooth flow of logistics distribution path network, the logistics distribution path optimization is designed to improve the efficiency of logistics distribution, a logistics distribution path optimization model is proposed based on artificial intelligence algorithm. A logistics distribution path search model based on rough set theory is established. Ant colony search method is used to design the artificial intelligence algorithm of logistics distribution path optimization. Adaptive weighting method is used to extract and schedule the information of logistics distribution path, and the shortest path optimization method is used to optimize the route planning of logistics distribution, which can reduce the path overhead and time cost of logistics distribution. The efficiency of logistics distribution is improved. The simulation results show that this method is used to construct the logistics distribution path model, which reduces the time cost and the road cost of the logistics distribution, and improves the throughput of the logistics distribution significantly.
Keywords: artificial intelligence algorithm; logistics distribution; ant colony search; shortest path optimization.
Study on fatigue of bus drivers based on biological model
by Xiao Hong
Abstract: In order to improve the fatigue detection ability of public transport drivers, the biometric modeling method is used to test and evaluate the fatigue of drivers. A fatigue detection model for public transport drivers is proposed based on biological mathematical model analysis, and the prevention and evaluation according to the fatigue test model is constructed. According to the visual, neural and blood supply characteristics of public transport drivers, the mathematical model of quantitative recursive statistical analysis of public transport drivers' biological fatigue is established by using descriptive statistical analysis method. The problem of public transport driver testing is transformed into an optimal solution problem for a continuous time-delay non-smooth system. Under the condition of non-smooth autonomous continuous boundedness, a biologic mathematical model of fatigue detection is constructed. The delay-dependent sufficient conditions of public transport driver fatigue testing are obtained to prevent and monitor driving fatigue accurately. The test results show that the model is accurate for public traffic drivers' fatigue testing. According to the biometric test results, it can reliably reflect whether the driver is tired or not, so that the danger alarm can be carried out and the driving safety can be ensured.
Keywords: biological mathematical model; public transport driver; fatigue test; vision; nerve.