This paper presents a biometric technique for identification of a person using the iris image. The iris is first segmented from the acquired image of an eye using an edge detection algorithm. The disk shaped area of the iris is transformed into a rectangular form. Described moments are extracted from the grayscale image which yields a feature vector containing scale, rotation, and translation invariant moments. Images are clustered using the k-means algorithm and centroids for each cluster are computed. An arbitrary image is assumed to belong to the cluster whose centroid is the nearest to the feature vector in terms of Euclidean distance computed. The described model exhibits an accuracy of 98.5%.

/pdf/iris-recognition-using-image-moments-and-k-means-algorithm-3k17s34nqq.pdf

Iris recognition using image moments and k-means algorithm.

In this paper, an efficient method based on a novel variation of local binary pattern (LBP), average local binary pattern (ALBP), is proposed for iris recognition, which is less sensitive to histogram equalization and parameters' selection and has low computation complexity. Center pixel and its neighborhood are the crucial elements involved in basic LBP. ALBP places high value on the significance of center pixel, while nearly all other variations of LBP have been focusing on the selection of neighborhood. Four candidates for the modification of the center pixel are elected and validated, respectively. In the proposed framework, the valid iris region firstly is localized and then normalized into a uniform rectangular. Then the normalized iris is chopped into several sub-images, and ALBP operator is applied to each sub-image to obtain individual histogram feature. Every histogram feature is then concatenated to form a global iris feature vector. Nearest neighbor classifier and support vector machine are employed to validate the recognition performance. Experimental results on CASIA-IrisV4 (including CASIA-Iris-Interval and CASIA-Iris-Thousand) and UBIRIS.V1 database show that our method achieves competitive recognition performance (optimal recognition rate is $$99.91\,\%$$99.91%) compared with other methods using the same databases.

Iris recognition based on a novel variation of local binary pattern

Although iris recognition has achieved big successes on biometric identification in recent years, difficulties in the collection of iris images with high resolution and in the segmentation of valid regions prevent it from applying to large-scale practical applications. In this paper, we present an eye recognition framework based on deep learning, which relaxes the data collection procedure, improves the anti-fake quality, and promotes the performance of biometric identification. Specifically, we propose and train a mixed convolutional and residual network (MiCoRe-Net) for the eye recognition task. Such an architecture inserts a convolutional layer between every two residual layers and takes the advantages from both of convolutional networks and residual networks. Experiment results show that the proposed approach achieves accuracies of 99.08% and 96.12% on the CASIA-Iris-IntervalV4 and the UBIRIS.v2 datasets, respectively, which outperforms other classical classifiers and deep neural networks with other architectures.

Eye Recognition With Mixed Convolutional and Residual Network (MiCoRe-Net)

2 Abstract: This article presents a biometric technique for identification of a person using the iris image. The iris is first segmented from the acquired image using an edge detection algorithm. The disk shaped area of iris is transformed into rectangular form. Since the image is in high resolution it will require a soaring number of neurons in the input layer, if the image data is directly clamped to the neural network, making the problem irrepressible. Resolution to this problem is sought through compression of data using a bottleneck network. The recognition neural network is trained using the compressed data using back propagation. The overall accuracy of system is 95.2 percent after reasonable training.

/pdf/a-neuro-cognitive-approach-for-iris-recognition-using-back-5183xew0b4.pdf

A Neuro-Cognitive Approach for Iris Recognition Using Back Propagation

Feature extraction for different distances of visible reflection iris using multiscale sparse representation of local Radon transform

We propose an improved iris recognition method to identify the person accurately by using a novel iris segmentation scheme based on the chain code and the collarette area localization. The collarette area is isolated as a personal identification pattern, which captures only the most important areas of iris complex structures, and a better recognition accuracy is achieved. The idea to use the collarette area is that it is less sensitive to the pupil dilation and usually not affected by the eyelids or the eyelashes. The deterministic feature sequence is extracted from the iris images using the 1D log-Gabor wavelet technique and used to train the support vector machine (SVM) as iris pattern classifiers. The parameters of SVM are tuned to improve the overall system performance. Our experimental results also indicate that the performance of SVM as a classifier is far better than the performance of backpropagation neural network (BPNN), K-nearest neighbor (KNN), Hamming distance and Mahalanobis distances. The proposed innovative technique is computationally effective as well as reliable in term of recognition rate of 99.56%.

R.C. Debnath

Papers

Multi-class SVM based iris recognition