No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Analysis of Iris Identification System by Using Hybrid Based PSO Classifier
Abstract
In the present days, Iris recognition as a physiological attribute of biometric is an important biometric process. Human eye iris acts as a significant task in vast identification of a human being. In this research work, the block sum and Haar transform algorithms i.e. hybrid algorithms are presented as a feature extraction method. After extracted features, hybrid based PSO classifier is used for classification. Hybrid based PSO classifier contains the combination of weighted DAG multi-class SVM and SNN i.e. spiking neural network. Internally weighted DAG multi class SVM is used for classification and SNN is used for optimization of PSO. For performing an experiment, we have taken 280 images of eye from 28 individuals and every person has 10 images of eye from CASIA version VI iris database. Experimental result shows that the hybrid PSO based classifier gives superior result in evaluation with other methods i.e. SVM and ANN. By using this method the average classification accuracy is 99.99%. The entire simulation is done on MATLAB R2013@ environment with tested images and accuracy as a parameter.
... Admovic et al. [33] proposed an approach for iris recognition by using stylometric features and random forest machine learning methods. The hybrid-based particle swarm optimization (PSO) is used as a classifier and proposed an iris recognition system by Gale et al. [34]. Hybrid-based PSO is a combination of a weighted directed acyclic graph (DAG) SVM and spiking neural networks (SNN). ...
Many government and commercial organizations are using biometric authentication systems instead of a password or token-based authentication systems. They are computationally expensive if more users are involved. To overcome this problem, a biometric system can be created and deployed in the cloud which then can be used as a biometric authentication service. Privacy is the major concern with cloud-based authentication services as biometric is irrevocable. Many biometric authentication systems based on cancelable biometrics are developed to solve the privacy concern in the past few years. But the existing methods fail to maintain the trade-off between speed, security, and accuracy. To overcome this, we present a multi-instance cancelable iris system (MICBTDL). MICBTDL uses a convolutional neural network trained using triplet loss for feature extraction and stores the feature vector as a cancelable template. Our system uses an artificial neural network as the comparator module instead of the similarity measures. Experiments are carried on IITD and MMU iris databases to check the effectiveness of MICBTDL. Experimental results demonstrate that MICBTDL accomplishes fair performance when compared to other existing works.
Multimodal fusion for biometrics recognition system had gained specific attention nowadays thanks to its remarkable valuable results. For this approach, classification methods have been the basis of important recognition accuracy improvements. The artificial neural networks (ANN) and support vector machines (SVM) belong to this class of methods. This paper presents comparison concerning the performances of the some methods that have been successfully applied to the fusion of scores for multimodal biometric recognition. After recognizing each single modality which was the fingerprint, the face as well as the voice, we recovered three similarity scores. These scores are then introduced into the classification system based on neural networks and on support vector machine techniques. Experimental results demonstrate that the identity established by such an integrated system is more reliable than the established identity by fingerprint recognition system, facial verification system and a voice verification system. Fusion phases are performed at score level. An average rate (= 57,69 %) is obtained by fusion with ANN. While fusion with the SVM gives an average rate equal to (= 63,31 %). A brief
In today's world the higher stable and distinct biometric characteristics to identify and / or to verify any person are the human iris. Iris recognition system consists image acquisition, localization, normalization, features extraction and matching. Iris images are taken from CASIA iris VI database for study. In this paper we make a comparative study of performance of image transform using Haar transform, PCA, Block sum algorithm and hybrid algorithm for iris verification to extract features on specific portion of the iris for improving the performance of an iris recognition system. The hybrid methods are evaluated by combining Haar transform and block sum algorithm. The classifiers used in this study are hybrid classifier i.e. ANN and FAR/FRR and the experimental results show that this technique produces good performance on CASIA VI iris database.
The randomness of iris pattern makes it one of the most reliable biometric traits. On the other hand, the complex iris image structure and the various sources of intra-class variations result in the difficulty of iris representation. Although, a number of iris recognition methods have been proposed, it has been found that several accurate iris recognition algorithms use multiscale techniques, which provide a well-suited representation for iris recognition. In this paper and after a thorough analysis and summarization, a multiscale edge detection approach has been employed as a pre-processing step to efficiently localize the iris followed by a new feature extraction technique which is based on a combination of some multiscale feature extraction techniques. This combination uses special Gabor filters and wavelet maxima components. Finally, a promising feature vector representation using moment invariants is proposed. This has resulted in a compact and efficient feature vector. In addition, a fast matching scheme based on exclusive OR operation to compute bits similarity is proposed where the result experimentation was carryout out using CASIA database. The experimental results have shown that the proposed system yields attractive performances and could be used for personal identification in an efficient and effective manner and comparable to the best iris recognition algorithm found in the current literature.
Iris pattern recognition using SVM and ANN
- A A Khan
- S Kumar
- M Khan
Study of two different methods for Iris recognition support vector machine and phase based method
- Kaur Gangapreet
- Kaur Dilpreet
- Singh Dheerendra
- Gangapreet Kaur
Performance analysis on iris feature extraction using PCA, Haar transform and block sum algorithm
- A G Gale
- S S Salankar
An optimize iris recognition system using MOGA followed by combined classifier
- P Punyani
- A Kumar
- R Gupta