Multimodal biometric system can be accomplished at different levels of fusion and achieve higher recognition performance than the unimodal system. This paper concerned to study the performance of different classification techniques and fusion rules in the context of unimodal and multimodal biometric systems based on the electrocardiogram (ECG) and fingerprint. The experiments are conducted on ECG and fingerprint databases to evaluate the performance of the proposed biometric systems. MIT-BIH database is utilized for ECG, FVC2004 database is utilized for the fingerprint, and further experiments are being performed to evaluate the proposed multimodal system with 47 subjects from virtual multimodal database. The performance of the proposed unimodal and multimodal biometric systems is measured using receiver operating characteristic (ROC) curve, AUC (area under the ROC curve), sensitivity, specificity, efficiency, standard error of the mean, and likelihood ratio. The findings indicate AUC up to 0.985 for sequential multimodal system, and up to 0.956 for parallel multimodal system, as compared to the unimodal systems that achieved AUC up to 0.951, and 0.866, for the ECG and fingerprint biometrics, respectively. The overall performance of the proposed multimodal systems is better than that of the unimodal systems based on different classifiers and different fusion levels and rules.

Multimodal biometric systems based on different fusion levels of ECG and fingerprint using different classifiers

this research proposed the fingerprint recognition by euclidean distance method. the system uses a technique of image processing, which consists of 3 major components, which are: 1) preprocessing component, the module that reduces the noise of the original images and adjusts the sharpness of the lined pattern that is the components of the fingerprint, 2) feature extraction component, the module that defines the position of the core point used as a reference point and finds out the position of the bifurcation points, and 3) fingerprint recognition component, the module to compare the shape context of training and testing data sets. based on the experimental results, the system has acceptable accuracy with average access time of around 19.68 seconds per image.

Fingerprint Recognition by Euclidean Distance

Traditional pattern recognition systems were implemented using neural network systems in order to recognize partial prints ranging from 100%, 75%, 50%, 40% and 30% of the whole fingerprint image. The uniqueness of the patterns of each individual fingerprint served as the motivation in pursuing this research in identifying partial fingerprints where the minutiae pattern identification method cannot be used due to incomplete information arising from the partiality of the fingerprint. Promising results are achieved during experimentations. This system is also able to identify partial prints subjected to incorporated noise and at most 10% blank spots within the prints. This partial fingerprint recognition system is intended to support forensic science particularly during crime scene investigations where full images of the fingerprints are impossible to locate.

Neural network based partial fingerprint recognition as support for forensics

Internet of Things refers to a paradigm consisting of a variety of uniquely identifiable day to day things communicating with one another to form a large scale dynamic network. Securing access to this network is a current challenging issue. This paper proposes an encryption system suitable to IoT features. In this system we integrated the fuzzy commitment scheme in DCT-based recognition method for fingerprint. To demonstrate the efficiency of our scheme, the obtained results are analyzed and compared with direct matching (without encryption) according to the most used criteria; FAR and FRR.

IoT Securing System using Fuzzy Commitment for DCT-based Fingerprint Recognition

Nowadays, biometric is a most technique to authenticate /identify human been, because its resistance against theft, loss or forgetfulness. However, biometric is subject to different transmission attacks. Today, the protection of the sensitive biometric information is a big challenge, especially in current wireless networks such as internet of things where the transmitted data is easy to sniffer. For that, this paper proposes an Eigens-Fingerprint-based biometric cryptosystem, where the biometric feature vectors are extracted by the Principal Component Analysis technique with an appropriate quantification. The key-binding principle incorporated with bit-wise and byte-wise correcting code is used for encrypting data and sharing key. Several recognition rates and computation time are used to evaluate the proposed system. The findings show that the proposed cryptosystem achieves a high security without decreasing the accuracy.

Eigen-Fingerprints-Based Remote Authentication Cryptosystem

In the fingerprint recognition application utilizing more information other than minutiae is much helpful. In this paper we proposed discrete cosine transform (DCT) based feature vector for fingerprint representation and matching. The transform is applied on the image directly without any preprocessing. By dividing the transformed image into various blocks, standard deviation is calculated for each block and such 96 standard deviations will form the feature vector. This feature vector is used in matching stage. Work is further extended by forming feature vector of 36 standard deviation from mid and high frequency bands of DCT coefficients. Total 8 images per person are captured for 15 individual and training set is prepared with the help of k images where k varies from 1 to 8. Results are checked against remaining images as well as newly captured image of same person. Results are represented in terms of Truly accepted, Falsely accepted and Rejected users. Recognition rate (%) are calculated with different threshold values. In identification mode 100% results are obtained for a typical threshold value.

M.K. Sahu

Papers

DCT feature based fingerprint recognition