Biometric Recognition Using Fusion
01 Jan 2020-pp 1320-1329
TL;DR: In this paper, a multimodal biometric system uses more than one biometric trait or modality for recognition of an individual, which fuses different types of input at different levels: Score level, Feature level and Decision level.
Abstract: Human identification systems based on biometrics are used in many applications to increase the security level. There are different biometric traits which are used in various applications. Monomodal biometric systems face many challenges such as error rates, using only single biometric for human recognition. Today, to increase the security of the authentication system, various multimodal biometric systems are proposed. A multimodal biometric system uses more than one biometric trait or modality for recognition of an individual. Multimodal biometric systems fuses different types of input at different level: Score level, Feature level and Decision level to get the better performance of the system.
Citations
More filters
Dissertation•
01 Jan 2012
TL;DR: In this article, the fusion of iris and fingerprint biometrics and their potential application as biometric identifiers is explored, and individual comparison scores obtained from the iris this article and fingerprints are combined at score-level using a three score normalization techniques (Min-Max, Z-Score, Hyperbolic Tangent) and four score fusion approaches (Minimum Score, Maximum Score Simple Sum and User Weighting).
Abstract: The majority of deployed biometric systems today use information from a single biometric technology for verification or identification. Large-scale biometric systems have to address additional demands such as larger population coverage and demographic diversity, varied deployment environment, and more demanding performance requirements. Today's single modality biometric systems are finding it difficult to meet these demands, and a solution is to integrate additional sources of information to strengthen the decision process. A multibiometric system combines information from multiple biometric traits, algorithms, sensors, and other components to make a recognition decision. Besides improving the accuracy, the fusion of biometrics has several advantages such as increasing population coverage, deterring spoofing activities and reducing enrolment failure. The last 5 years have seen an exponential growth in research and commercialization activities in this area, and this trend is likely to continue. Therefore, here we propose a novel multimodal biometric authentication approach fusing iris and fingerprint traits at score-level. We principally explore the fusion of iris and fingerprint biometrics and their potential application as biometric identifiers. The individual comparison scores obtained from the iris and fingerprints are combined at score-level using a three score normalization techniques (Min-Max, Z-Score, Hyperbolic Tangent) and four score fusion approaches (Minimum Score, Maximum Score Simple Sum and User Weighting). The fused-score is utilized to classify an unknown user into the genuine or impostor.
18 citations
TL;DR: The aim of this paper is to detect impostors using various machine learning techniques to see which combination works best for speaker recognition and classification, and to examine the 7 classifiers on two datasets, the extent of accuracy achieved for each classifier.
Abstract: Voice is a Special metric that, in addition to being natural to users, offers similar, if not higher, levels of security when compared to some traditional biometrics systems. The aim of this paper is to detect impostors using various machine learning techniques to see which combination works best for speaker recognition and classification. We present several methods of audio preprocessing, such as noise reduction and vocal enhancements, to improve the audios available in real environments. Mel Frequency Cepstral Coefficients (MFCC) are extracted for each audio, along with their differentials and accelerations, to verify machine learning classification methods such as PART, JRip, Nave Bayes, RT, J48, Random Forest, and k-Nearest Neighbor Classifiers. examine the 7 classifiers on two datasets, the extent of accuracy achieved for each classifier. Among the high performance were the random forest algorithm and the naive bias algorithm, and the weak performance of the PART algorithm.
13 citations
TL;DR: In this paper, the authors use a direct spread spectrum based on CDMA, especially mobile wireless third generation with a wide bandwidth that can send large data such as multimedia between patients and doctors at anytime and anywhere.
Abstract: When medical technology, communication, and computers were developed and put into practice in the treatment and follow-up of remote health care patients, the difficulty of communicating with the medical team or patients was due to bandwidth limitations. Sending a significant volume of data between medical personnel and patients, as well as power usage, is not permitted. Wireless sensors use a direct spread spectrum based on CDMA, especially mobile wireless third generation with a wide bandwidth that can send large data such as multimedia between patients and doctors at anytime and anywhere. The wireless sensor network decreases the power consumption of the transmitter sensor and the receiver sensor by using binary transmission and multipath. The accuracy of the measurement depends on the bandwidth of the unit to calculate the time of arrival (TOA). The TOA forecasts come and go together to pass them to the receiver because of the delay of the MPC and the reduction of replicas of the initial signal. The diffusion time caused by the moving signal through barriers, adding a positive bias to the TOA.
7 citations
TL;DR: In this paper, a downdraft gasifier of 1 MW for electricity purposes was designed, built, and commissioned for developing countries as essential electricity and green energy source, where wood pellets were the feedstock at a feeding rate of 300 kg/hr.
Abstract: This work aims to design, build, and commission a downdraft gasifier of 1 MW for electricity purposes. Wood pellets were the feedstock at a feeding rate of 300 kg/hr. A reaction temperature's range from 700⁰C to 974⁰C was reached by controlling the equivalence ratio. Replacing the airflow with exhaust gasses led to getting a temperature of (620–850 ⁰C), using wet biomass instead of the drying one led to having a reaction temperature of 830–1070 ⁰C, which is almost the same in the case of air and that of the exhaust gas. The temperature of the produced syngas at the outlet was found between 180⁰C to 220⁰C. The analysis of the produced syngas shows that carbon monoxide: 14.4–19.2%, hydrogen 16–20%, carbon dioxide 7.1–11.2%, and a small volume of methane 2–3%. Such built downdraft gasifiers could be a prospective system for developing countries as essential electricity and green energy source.
5 citations
TL;DR: This study aims the impostor is a very cunning person who reaches an obsessive stage to perfection in impersonating someone in actual life, concentrates on his biometric.
Abstract: This study aims the impostor is a very cunning person who reaches an obsessive stage to perfection in impersonating someone in actual life, concentrates on his biometric. He analyzes the controls, restrictions, and obstacles that he will face to overcome them. The technologies biometric recognition performs a greatly important role in impostor detection. Biometrics properties refer to the automatic recognition of persons depending on their behavioral and physiological characteristics. Biometrics comprises face recognition, fingerprint, voice recognition, retinal scanning, and so on. Biometrics may increment the reliability of an ID card system. In this paper, a review of the concepts mentioned above will be provided. At first, a presentation about a procedural overview of biometric recognition technologies, ID card systems. Then dissection will be presented for the review of the most recent techniques. A description of each concept will be given and a comparison study is achieved with formal discussion and analysis for each approach result introduces in this study. Finally, a summary of the research results is given.
4 citations
References
More filters
12 Dec 2007
TL;DR: The aim of this paper is to study the fusion at feature extraction level for face and fingerprint biometrics by extracting independent feature pointsets from the two modalities, and making the two pointsets compatible for concatenation.
Abstract: The aim of this paper is to study the fusion at feature extraction level for face and fingerprint biometrics. The proposed approach is based on the fusion of the two traits by extracting independent feature pointsets from the two modalities, and making the two pointsets compatible for concatenation. Moreover, to handle the 'problem of curse of dimensionality', the feature pointsets are properly reduced in dimension. Different feature reduction techniques are implemented, prior and after the feature pointsets fusion, and the results are duly recorded. The fused feature pointset for the database and the query face and fingerprint images are matched using techniques based on either the point pattern matching, or the Delaunay triangulation. Comparative experiments are conducted on chimeric and real databases, to assess the actual advantage of the fusion performed at the feature extraction level, in comparison to the matching score level.
164 citations
01 Nov 2010
TL;DR: The ATVS-UAM fusion approach submitted to the quality-based evaluation of the 2007 BioSecure Multimodal Evaluation Campaign, whose aim was to compare fusion algorithms when biometric signals were generated using several biometric devices in mismatched conditions is described and evaluated.
Abstract: As biometric technology is increasingly deployed, it will be common to replace parts of operational systems with newer designs. The cost and inconvenience of reacquiring enrolled users when a new vendor solution is incorporated makes this approach difficult and many applications will require to deal with information from different sources regularly. These interoperability problems can dramatically affect the performance of biometric systems and thus, they need to be overcome. Here, we describe and evaluate the ATVS-UAM fusion approach submitted to the quality-based evaluation of the 2007 BioSecure Multimodal Evaluation Campaign, whose aim was to compare fusion algorithms when biometric signals were generated using several biometric devices in mismatched conditions. Quality measures from the raw biometric data are available to allow system adjustment to changing quality conditions due to device changes. This system adjustment is referred to as quality-based conditional processing. The proposed fusion approach is based on linear logistic regression, in which fused scores tend to be log-likelihood-ratios. This allows the easy and efficient combination of matching scores from different devices assuming low dependence among modalities. In our system, quality information is used to switch between different system modules depending on the data source (the sensor in our case) and to reject channels with low quality data during the fusion. We compare our fusion approach to a set of rule-based fusion schemes over normalized scores. Results show that the proposed approach outperforms all the rule-based fusion schemes. We also show that with the quality-based channel rejection scheme, an overall improvement of 25% in the equal error rate is obtained.
69 citations
27 Oct 2009
TL;DR: The proposed multimodal biometric system overcomes the limitations of individual biometric systems and also meets the response time as well as the accuracy requirements.
Abstract: This paper presents a multimodal biometric recognition system integrating palmprint, fingerprint and face based on score level fusion. The feature vectors are extracted independently from the pre-processed images of palmprint, fingerprint and face. The feature vectors of query images are then compared individually with the enrollment templates which are taken and stored during database preparation for each biometric trait respectively. The individual matching scores generated after matching of query images with database images are passed to the fusion module. Fusion module performs score normalization and fusion of normalized scores by weighted sum rule. Weights associated with each biometric trait for a specific user indicates the importance of corresponding biometric characteristic possessed by the user. These individual normalized scores along with their weights are finally combined into a total score by sum rule, which is passed to the decision module which declares the person as genuine or an imposter. The identity established by this system is more reliable than the identity established by individual biometric systems. Integrating multiple biometric traits improves recognition performance and reduces fraudulent access. The proposed multimodal biometric system overcomes the limitations of individual biometric systems and also meets the response time as well as the accuracy requirements.
56 citations
10 Jun 2012
TL;DR: FP and FKP are integrated in order to construct an efficient multi-biometric recognition system based on matching score level and image level fusion, and the experimental results showed that the designed system achieves an excellent recognition rate on the Hong Kong polytechnic university (PolyU) FKp and high resolution fingerprint database.
Abstract: Biometrics is an effective technology for personnel identity recognition, but uni-modal biometric systems which use a single trait for recognition will suffer from problems like noisy sensor data, non-universality, lack of distinctiveness of the biometric trait, and spoof attacks. These problems can be tackled by using multi-biometrics in the system. Hand-based person recognition provides a reliable, low-cost and user-friendly viable solution for a range of access control applications. As one of the most popular biometric traits, fingerprints (FP) are widely used in personal recognition. However, a novel hand-based biometric feature, Finger-Knuckle-Print (FKP), has attracted an increasing amount of attention. In this paper, FP and FKP are integrated in order to construct an efficient multi-biometric recognition system based on matching score level and image level fusion. In this study we use the minimum average correlation energy (MACE) and Unconstrained MACE (UMACE) filters in conjunction with two correlation plane performance measures, max peak value and peak-to-sidelobe ratio, to determine the effectiveness of this method. The experimental results showed that the designed system achieves an excellent recognition rate on the Hong Kong polytechnic university (PolyU) FKP and high resolution fingerprint database.
23 citations
01 Jan 2017
TL;DR: The results show that face recognition achieved a high performance with deep learning features under different light and expression conditions, however, multi-biometric results have reached higher genuine match rate (GMR) performance and lower false acceptance rate (FAR).
Abstract: Nowadays, with the increasing use of biometric data, it is expected that systems work robustly and they can give successful results against difficult situations and forgery. In face recognition systems, variables such as direction of light, facial expression and reflection makes identification difficult. With biometric fusion, both safe and high performance results can be achieved. In this work, Eurocom Kinect Face dataset and BodyLogin Gesture Silhouettes dataset are used to create a virtual dataset and they were fused with score level. For face database, VGG Face deep learning model was used as feature extractor and energy imaging method was used for extracting gesture features. Afterwards the features reduced by principal component analysis and similarity scores were produced with standard deviation Euclidean distance. The results show that face recognition achieved a high performance with deep learning features under different light and expression conditions, however, multi-biometric results have reached higher genuine match rate (GMR) performance and lower false acceptance rate (FAR). As a result of this process, gesture energy imaging can be used for person recognition and for multi biometric data.
22 citations