Features Subset Selection using Improved Teaching Learning based Optimisation (ITLBO) Algorithms for Iris Recognition

The success of iris recognition depends mainly on two factors: image acquisition and an iris recognition algorithm. In this study, we present a system that considers both factors and focuses on the latter. The proposed algorithm aims to find out the most efficient wavelet family and its coefficients for encoding the iris template of the experiment samples. The algorithm implemented in software performs segmentation, normalization, feature encoding, data storage, and matching. By using the Haar and Biorthogonal wavelet families at various levels feature encoding is performed by decomposing the normalized iris image. The vertical coefficient is encoded into the iris template and is stored in the database. The performance of the system is evaluated by using the number of degrees of freedom, False Reject Rate (FRR), False Accept Rate (FAR), and Equal Error Rate (EER) and the metrics show that the proposed algorithm can be employed for an iris recognition system.

Iris recognition is one of the emerging areas as the demand for security in social and personal areas is increasing day by day. The most challenging step in the process of iris recognition is accurate iris localization. As it significantly affect the further processing of feature extraction and template matching stages. Traditional algorithms work efficiently and accurately in localizing iris from eye images taken in constrained environmental conditions. But their accuracy gets affected when eye images are taken in unconstrained environmental conditions. The proposed algorithm starts with determining the exact location of iris even in the presence of specular highlights and non-ideal environmental conditions. It also works well for noisy iris images, and the accuracy of segmentation stage has increased when intuitionistic fuzzy is applied on the UBIRIS v2 iris images.

This study focuses on iris recognition using deep learning techniques. The EfficientDet network model was employed for both iris detection and recognition tasks. Four datasets were utilized to train and evaluate the deep learning network. The model was trained to extract iris features and classify individuals based on their unique iris patterns. The proposed method achieved a high recognition rate of over 98% across multiple dataset evaluations. For real‐time implementation on an embedded system, the trained model was quantized to an 8‐bit integer format to accommodate resource‐constrained devices. Despite this quantization, the recognition accuracy remained high, reaching 97%. By incorporating an Edge TPU accelerator alongside a Raspberry Pi system, the processing speed reached up to 10 frames per second during real‐time iris camera testing, demonstrating the feasibility of real‐time iris recognition. An intruder test was conducted to assess the system's robustness in preventing unauthorized access. The False Acceptance Rate (FAR) was measured to assess the likelihood of incorrectly accepting an unauthorized individual. Experimental results show that the FAR can be reduced to zero by applying additional temporal constraints, effectively preventing unauthorized individuals from passing the iris recognition‐based access control system.

Biometric is an automated recognition process that measures the unique human characteristics, such as iris recognition. Of late, different kinds of iris recognition techniques are available that can be used in different platforms. Daugman algorithm is one of the iris recognition techniques that provide high percentage of accuracy. This study aims to develop a device that performs Iris Recognition using Daugman algorithm on Raspberry Pi. Through the implementation of Daugman algorithm on Raspberry Pi, more applications may come up for iris recognition. Also to make the system adapt with the limitations of Raspberry Pi and the existing application, this work created an image processing function that preprocesses the image before it is passed to the application. Based on the tests implemented, the False Acceptance Rate (FAR) and False Rejection Rate (FRR), the system is mostly to reject registered actors than to accept impostors due to the hardware limitations.

This chapter aims to enhance the security of textual passwords by adding a new layer that involves an individual's typing rhythm without requiring additional devices. It discusses authentication methods and develops a textual password authentication system, which takes advantage of the fact that every user has a distinct way of typing. Twenty participants, including novice and expert users, assessed the proposed system. The measuring metrics used were the False Acceptance Rate (FAR) and False Rejection Rate (FRR). The flight time and typing rhythm were the biometric identification template. The results indicated low false rejection and acceptance rates, with the initial session registering 0.0 FRR and 0.1 FAR. Two weeks later, the second session recorded 0.25 FRR and 0.0 FAR. Using this model, users do not need to select a complex password they might forget. Instead, they can utilize a good rhythm different from their natural typing rhythm, making it challenging to guess.

The increasing reliance on Automated Teller Machines (ATMs) has highlighted the urgent need for advanced authentication mechanisms to safeguard user transactions against fraud and unauthorized access. However, traditional methods such as Personal Identification Numbers (PINs) and passwords remain vulnerable to attacks, while unimodal biometric systems face challenges of inter-class variance, environmental interference, and non-universality. Specifically, single-modality approaches and conventional ATM cameras fall short in capturing reliable biometric features under varying conditions, while the effectiveness of bimodal approaches in such environments has not been adequately investigated. Therefore, this study developed a bimodal biometric authentication system integrating face and iris recognition with Gray Level Co-Occurrence Matrix (GLCM) for enhanced ATM security. The system leverages GLCM for powerful texture feature extraction from both modalities, capturing intricate spatial relationships that are difficult to spoof. The extracted feature vectors were used to train Support Vector Machines (SVM) with a Radial Basis Function (RBF) kernel as classifiers for both face and iris recognition. The final authentication decision was made using Boolean OR rule fusion. The system achieved a remarkable accuracy of 98.2%, with a False Acceptance Rate (FAR) of 1.8% and a False Rejection Rate (FRR) of 1.2%. These results demonstrably outperformed comparable unimodal systems and existing biometric ATMs, validating the proposed framework as a highly secure and efficient solution for financial authentication systems.

The Iris recognition is commonly used as a security system due to its robustness against imposters. Iris datasets are huge and hence those datasets occupy more space. Iris image compression has become an important part of better performance like speed and data storage. The portable Iris system is in huge demand. That portable systems need to transmit the iris images through a very small bandwidth channel. To reduce the time for transferring a huge number of data over small bandwidth channel, iris file can be compressed to some extent to minimize the size. Another problem is that when an image is captured, it captures some noise that disturbs the recognition performance, so, denoising is required for noise-free images. This paper separately analyzes the impact of wavelet compression along with denoising on iris images. The compression analysis is done using Embedded Zero Tree Wavelet, the other technique used is Set Partitioning in Hierarchical Tree and the third technique used is Spatial-Orientation Tree Wavelet. Denoising is done using different wavelets Daubechies, Haar, Biorthogonal and Fejer-Korovkin. The impact of the wavelet compression and denoising techniques on recognition performance are compared with False Rejection Rate and False Acceptance Rate. The quality of a compressed image is calculated with different quality metrics. This work establishes that compression and denoising of the images minimally affect the recognition performance.

This paper proposes an authentication protocol based on post-quantum cryptography and iris recognition. The process uses a modified version of the original McEliece cryptographic algorithm that provides post-quantum security. Additionally, iris recognition is included in the decryption process, ensuring that only authentic users have access to confidential information, or can operate critical infrastructures or systems that must be protected against malicious users or impostors. A distributed framework, in which several autonomous entities interact with each other, participates in the whole process to reduce the odds of a successful attack. Biometric templates are encrypted and safely stored in a database. These templates meet all requirements described in the ISO/IEC 24745 standard regarding data protection. Due to the homomorphic property of McEliece, the decryption process operates using the encrypted version of these biometric templates, ensuring that critical information is not revealed. The use of McEliece does not affect the overall accuracy of the recognition process, so that the same False Acceptance and False Rejection rates are obtained when compared with the non-protected scheme.

We examined the false acceptance rate (FAR) and false rejection rate (FRR) of varying precision verification experimental designs. Analysis of variance was applied to derive the subcomponents of imprecision (ie, repeatability, between-run, between-day imprecision) for complex matrix experimental designs (day × run × replicate; day × run). For simple nonmatrix designs (1 day × multiple replicates or multiday × 1 replicate), ordinary standard deviations were calculated. The FAR and FRR in these different scenarios were estimated. The FRR increased as more samples were included in the precision experiment. The application of an upper verification limit, which seeks to cap FRR at 5% for multiple experiments, significantly increased the FAR. The FRR decreases as the observed imprecision increases relative to the claimed imprecision and when a greater number of days, runs, or replicates are included in the verification design. Increasing the number or days, runs, or replicates also reduces the FAR for between-day imprecision and repeatability. Design of verification experiments should incorporate the local availability of resources and analytical expertise. The largest imprecision component should be targeted with a greater number of measurements. Consideration of both FAR and FRR should be given when committing a platform into service.

It is widely assumed that, barring traumatic injury to the eye or eye surgery, a person's iris does not change over time. This implies that iris recognition performance does not degrade as time increases after initial enrollment-also known as the template-aging problem. We present results of the first study to investigate the validity of this assumption. We explore the effects of time lapse since enrollment on iris biometric match scores using a data set with four years time lapse between the earliest and most recent images of an iris (23 subjects, 46 irises, 6,814 total images). Experimental results are reported for three iris recognition algorithms. We find that the empirical mean of the distribution for similarity scores between iris images of the same person changes over four years-in biometric parlance, the match or genuine distribution changes. The change is statistically significant when comparing the mean of pairs of images taken within 120 days (4 months) and over 1200 days (3.25 years) for the three algorithms in our study. The mean of the match distribution changes so that the expected performance would degrade due to an increase in the false reject rate. Our results suggest that iris biometric templates undergo aging.

In this thesis we propose a set of algorithms to reduce the false rejection rate of iris recognition. Even though high recognition accuracy is claimed for iris recognition algorithms, high false rejection rates cause the impediment in worldwide use of iris biometrics.;A novel iris segmentation algorithm for non-ideal iris images treating iris as an elliptical object is proposed. Further, quality of the extracted iris image is improved using SVM based enhancement algorithm. In this algorithm, selected enhancement algorithms globally enhance the iris image and the learning algorithm synergistically fuses local information from these intermediate enhanced images. 1D log polar Gabor wavelet is then used to extract the textural features from the enhanced iris image and Euler numbers are used to extract the topological features. The extracted textural features give a global description of the iris image whereas the topological features are rotation, translation and scaling invariant. These two features are fused using the proposed match score and decision fusion algorithms. Among the three proposed fusion algorithm, SVM learning based match score fusion algorithm outperforms other fusion algorithms. Using CASIA, Miles, UBIRIS and UPOL iris databases, experimental results show that the proposed algorithm gives reduced failure to enroll rate with comparable accuracy.

Problem statement: In any real time biometric system processing speed and recognition time are crucial parameters. Reducing processing time involves many parameters like normalization, FAR, FRR, management of eyelid and eyelash occlusions, size of signature etc. Normalization consumes substantial amount of time of the system. This study contributes for improved iris recognition system with reduced processing time, False Acceptance Rate (FAR) and False Rejection Rate (FRR). Approach: To improve system performance and reliability of a biometric system. It avoided the iris normalization process used traditionally in iris recognition systems. The technique proposed here used different masks to filter out iris image from an eye. Comparative study of different masks was done and optimized mask is proposed. The experiment was carried on CASIA database consisting of 756 iris images of 108 persons. Each person contributes seven images of eye (108×7 = 756) images in the database. Results: In the proposed method: (1) Normalization step is avoided; (2) Computational time is reduced by 0.3342 sec; (3) Iris signature size is reduced; (4) Improved performance parameters. (With reduced feature size, proposed method achieves 99.4866% accuracy, 0.0069% FAR, 1.0198% FRR and significant increase in speed of the system). Conclusion: Iris signature proposed was comparatively small just of 1×24 size. Though Daugman’s method gives best accuracy of 99.90% but the iris signature length used by that algorithm is comparatively very high that is 1×2048 phase vector. Also Daugman has used phase information in signature formation. Our method gives a accuracy of 99.474% with a signature of comparatively very small length. This has definitely contributed to improve the speed.

The problem of speaker recognition from a given set of speakers for any language and any context is considered. A database of Russian numerals that contains speech segments from 216 men and 177 women, each of whom spoke from 400 to 800 words, is used for recognition. Speech has been recorded on different types of microphones in different rooms at the natural noise level. Recognition is based on solutions of the inverse problem of finding the voice excitation pulse shape for each pitch period by the known speech segment. The pulse shape is defined as the inverse Fourier transform of the regularized ratio of speech signal spectra at the intervals of the open and closed glottis. Recognition is carried out by ten parameters: the pitch period, the open glottis interval duration, times when the source amplitude is maximum, minimum, or zero, the amplitude ratio for the minimum and maximum source pulses, three decomposition ratios of the source function by the principal component method, and the vowel duration. In such a recognition procedure, in the case of the utterance of a word that contains one vowel, the false reject rate (FRR) for men is 1.7---5.4%, and the false acceptance rate (FAR) is 5.4---7.1%. For women FRR = 2---5.2% and FAR = 5.2---6.3%. The recognition error decreases with an increasing number of vowels in the speech signal. At 10 vowels, for men FRR = 0.05---0.2% and FAR = 0.07---0.8%, and for women FRR = 0.09---0.2% and FAR = 0.17---2.1%.

Among all biometric technologies, iris recognition is the most accurate and high confidence authentication system. Due to the limitations in PCA-based system, modified principal component analysis (PCA)-based feature extraction is proposed in iris recognition system. In the proposed system, features are extracted using Gaussian kernel PCA (GKPCA) and classified using support vector machine (SVM). GKPCA and SVM algorithms are evaluated using CASIA V3 iris database. The performances are compared with the existing PCA-based system. The proposed system achieves 96.67% of accuracy for 256 features using GKPCA linear SVM. False acceptance rate (FAR) and false rejection rate (FRR) are 0 and 3 respectively for linear SVM. The results show that the proposed system performs accurate localisation of patterns even in non-ideal conditions.

A Multi-biometric System amalgamates the evidences collected from the multiple sources or single source for person recognition based on templates like fingerprint, palm print and iris. These evidences can be combined at various levels like pixel level, score level, feature level, and decision level. A rich set of information is present at feature level, but there is a problem of dimensionality which affects the recognition system. This study focuses on a dimensionality reduction technique Principal Component Analysis (PCA), which reduces the feature vector after feature level fusion in three different multi-biometric systems. The Pre and Post analysis of reduction compares the size of feature vector and recognition time per template in the following systems. Fingerprint based recognition where two evidences like bifurcations and ridge endings are fused, palmprint recognition where 2D-Gabor and Log-Gabor features are concatenated, and Iris Recognition in which Log-Gabor and 2D-Gabor features are integrated. And also Performance of the system is evaluated Based on False Acceptance Rate and False Rejection Rate. The study has been carried out on FVC 2004 fingerprint, IIT Delhi palmprint, and IIT Delhi Iris datasets. The results have shown that a drastic improvement in the multimodal system performance after reduction in all metrics.