Fast learning for multibiometrics systems using genetic algorithms

Giot, Romain; El-Abed, Mohamad; Rosenberger, Christophe

doi:10.1109/hpcs.2010.5547127

Cited by 7 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the min rule which returns the minimum score value, (b) the mul rule which returns the product of all the scores, (c) the sum rule which returns the sum of the scores, (c) the weight rule which returns a weighted sum, and (d) an SVM implementation. The weighs of the weighted sum have been configured by using genetic algorithm on the training sets [50,51] (in order to give the best results as possible). The fitness function is the value of the EER and the genetic algorithm engine must lower this value.…”

Section: Resultsmentioning

confidence: 99%

Genetic programming for multibiometrics

Giot

Rosenberger

2012

Expert Systems with Applications

Self Cite

View full text Add to dashboard Cite

Biometric systems suffer from some drawbacks: a biometric system can provide in general good performances except with some individuals as its performance depends highly on the quality of the capture... One solution to solve some of these problems is to use multibiometrics where different biometric systems are combined together (multiple captures of the same biometric modality, multiple feature extraction algorithms, multiple biometric modalities. . . ). In this paper, we are interested in score level fusion functions application (i.e., we use a multibiometric authentication scheme which accept or deny the claimant for using an application). In the state of the art, the weighted sum of scores (which is a linear classifier) and the use of an SVM (which is a non linear classifier) provided by different biometric systems provid one of the best performances. We present a new method based on the use of genetic programming giving similar or better performances (depending on the complexity of the database). We derive a score fusion function by assembling some classical primitives functions (+, * , −, ...). We have validated the proposed method on three significant biometric benchmark datasets from the state of the art. Every day, new evolutions are brought in the biometric field of research. 3 These evolutions include the proposition of new algorithms with better per-4 formances, new approaches (cancelable biometrics, soft biometrics, ...) and 5 even new biometric modalities (like finger knuckle recognition [1], for example). 6 There are many different biometric modalites, each classified among three main 7 families (even if we can find a more precise topology in the literature) : 8 • biological : recognition based on the analysis of biological data linked to an 9 individual (e.g., DNA analysis [2], the odor [3], the analysis of the blood 10 of different physiological signals, as well as heart beat or EEG [4]); 11 • behavioural : based on the analysis of an individual behaviour while he is 12 performing a specific task (e.g., keystroke dynamics [5], online handwrit-13 ten signature [6], the way of using the mouse of the computer [7], voice 14 recognition [8], gait dynamics (way of walking) [9] or way of driving [10]); 15 • morphological based on the recognition of different particular physical pat-16 terns, which are, for most people, permanent and unique (e.g., face recog-17 nition [11], fingerprint recognition [12], hand shape recognition [13], or 18 blood vessel [14], ...). 19Nevertheless, there will always be some users for which a biometric modality 20 (or method applied to this modality) gives bad results, whereas, they are better 21 in average. These low performances can be implied by different facts: the quality 22 of the capture, the instant of acquisition and the individual itself but they have 23 the same implication (impostors can be accepted or user need to authenticate 24 themselves several times on the system before being accepted). Multibiometrics 25 allow to solve this problem while obtaining better perf...

show abstract

Section: Resultsmentioning

confidence: 99%

Genetic programming for multibiometrics

Giot

Rosenberger

2012

Expert Systems with Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kumar et al [14] proposed an ant colony optimization (ACO) based fuzzy binary decision tree for bimodal hand knuckle verification system, in which they uses ACO to choose optimal fusion parameter for each level of security.FBDT (fuzzy binary decision tree) are used for decision making purpose and classify the classes as genuine or imposter using matching score obtained from knuckle database. The application of GA (genetic algorithm) for selection of different fusion parameter at score level fusion was proposed by Romain Giot et al [15] in which they define a fitness function based on a fast Error Equal Rate computing method. They have tested three different kinds of score fusion methods whose parameter are automatically set by genetic algorithm, the score fusion functions have been validate on three different multibiometric database on which two are real and one is chimerical.Cherifi Dalila and Hafnaoui Imane [16] proposed a multibiometric system using combination of GA and PSO based on score level fusion, in which PSO and GA is applied to find the optimum weights associated to the modalities being fused.…”

Section: Related Workmentioning

confidence: 99%

Enhancing Performance of Multibiometric System using Ant Colony Optimization based on Score Level Fusion

Modak¹,

Jha²

2017

IJCA

View full text Add to dashboard Cite

Unimodal biometric systems have several inherent problems such as intra-class variation, noisy-sensor data, spoofing attacks and non-universality. To overcome this limitation multibiometric is a good option where we can use two or more individual modalities. In this paper we propose a multibiometric system to enhance the performance and minimize the error rate using Ant colony optimization (ACO) based on score level fusion. This work extracts the feature from two different modalities namely face and iris (left/right). In this work we use ACO as an optimization technique to select the fusion parameter like weights for the different biometric matcher and fusion rule which is used further for score level fusion. The experimental results show that the multibiometric system using ACO based on sum rule is outperform than the other fusion rule like product, tanh and exponential sum.

show abstract

“…Basically, three main families exist in biometric systems (a) Biological (b) Morphological (c) Behavioural [1].Every single biometric system has its own pros and cons. In order to make the system more reliable and accurate various modalities are combined together which lead to the concept of multimodal biometric systems.…”

Section: Introduction 11 What Is Biometrics?mentioning

confidence: 99%

A Novel Biometric System based on Hybrid Fusion Speech, Signature and Tongue

Kaur¹,

Singh²

2015

IJCA

View full text Add to dashboard Cite

There is urgent need to make behavioural biometric systems more reliable, robust and accurate. This paper presents feature level fusion of offline signature and speech and tongue biometrics. The fusion of speech and signature with tongue made the system more efficient. MFCC is used for feature extraction in speech and in signature DCT is used which is widely used for image and texture feature extraction. While the features of tongue are extracted using SIFT algorithm. Hybrid weighted average using Apirori two item set is applied for the fusion of extracted features of all the modalities. SVC2004 signature database are used for experimental results. Tongue database was collected by capturing pictures using digital camera while for speech CMU_ARCTIC database which is available at Language Technologies Institute at Carnegie Mellon University is used in the work. Features for both noisy and non-noisy samples have been collected separately. Gaussian noise was added to the system to check the performance in noisy environment. The proposed system also works efficiently on filtered noisy modalities with accuracy of 80%. The accuracy of the noiseless system is 88.75% with 0.06 % of FAR and with 0.05% of FRR and the noisy system have FAR 0.05% and FRR 0.15%. The ROC curves are calculated.

show abstract

Fast learning for multibiometrics systems using genetic algorithms

Cited by 7 publications

References 16 publications

Genetic programming for multibiometrics

Genetic programming for multibiometrics

Enhancing Performance of Multibiometric System using Ant Colony Optimization based on Score Level Fusion

A Novel Biometric System based on Hybrid Fusion Speech, Signature and Tongue

Contact Info

Product

Resources

About