No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Developing Iris Recognition System Based on Enhanced Normalization
Human body is the magical creation of god. It carries many interconnected systems. Changes in one system replicate changes in another system. Due to such interconnection of systems, we can analyze one system by observing changes in another system. Iridology supports the same theory. Iridology tells the relation between iris and other systems present in the body. Evaluation can be done in the form of the iris that speaks about the physical condition of different body parts. In the proposed method by observing different iris images without doing any complicated and time-consuming test, we can perform diagnosis for the brain tumour. This method can be used as a pre diagnosis tool.
The Biometric recognition is the study of identifying individuals based on their unique physiological or behavioral characteristics, includes iris, face, fingerprint, retina, vein, hand geometry, hand writing, human gait, signature, keystrokes and voice. Among the biometrics, an iris has unique structure and it remains stable over a person life time. So that iris recognition is regarded as the most accurate and reliable biometric recognition system.
In this paper, we proposed a technique that uses Principal Component Analysis (PCA) based on Discrete Wavelet Transformation (DWT) for selecting feature of iris templates to increase the efficiency of iris recognition. Basically, the idea of DWT is to convert the iris image into four frequency band. We are using one frequency band instead of four and applying PCA for further feature extraction. Experiments with iris images from the CASIA database present good results, showing that the proposed combination strategy of feature extraction is suitable
for increasing accuracy of iris recognition.
The use of human biometrics for automatic identity verification has become widespread. Mostly used human biometrics are face, fingerprint, iris, gait, retina, voice, hand geometry etc. Among them iris is an externally visible, yet protected organ whose unique epigenetic pattern remains stable throughout one's whole life. These characteristics make it very attractive to use as a biometric for identifying individuals. This paper presents a detailed study of iris recognition technique. It encompasses an analysis of the reliability and the accuracy of iris as a biometric of person identification. The main phases of iris recognition are segmentation, normalization, feature encoding and matching. In this work automatic segmentation is performed using circular Hough transform method. Daugman's rubber sheet model is used in normalization process. Four level phase quantization based 1D Log-Gabor filters are used to encode the unique features of iris into binary template. And finally the Hamming distance is considered to examine the affinity of two templates in matching stage. We have experimented a better recognition result for CASIA-iris-v4 database.
Iris texture is a natural password that has great advantages such as variability, stability, unique features for each person, and its importance in the security field. This makes an iris recognition system upper of other biometric methods used for human identification. Recent science is interested to develop intelligent systems able to identify persons based on the texture of their iris. We proposed a new feature extraction method based on local and directional texture information. The proposed feature extraction method gets both local and global relevant information and faster than commonly used method. In the experimental parts, our system is compared to other famous and recent iris recognition systems using CASIA iris dataset. Experiments demonstrate that the proposed system gives better recognition rate (99.96 %) compared to other systems.
Abstract: :Iris recognition becomes one of the most accurate and secures biometric method used today. The execution time of the iris recognition algorithm on general purpose sequential system as central processing unit is too high, so it cannot work in the real time applications. In this paper, an enhancement for the iris recognition system was applied for each processing part to speed up the execution time and make the opportunity to work in real time applications.
Two enhancements were made in this paper, the first one by using hardware implementation for all the iris recognition process which are: Segmentation, Normalization, Feature extraction, and Hamming distance using the FPGA. The second enhancement is by choosing a small part (quarter) from the iris region which contains sufficient features to make the recognition, hence reducing the processing time.
Multimodal biometric systems are being adopted as the most effective solution for security breaches these days as they are more reliable and accurate than unimodal systems. These are the pattern recognition systems which are used for identification/verification of the person using their physical or behavioral traits. Background removal for extracting palm print image from an unconstrained background is done using FCM (fuzzy c-mean) technique. Texture feature extraction of both iris and palm print is done using Ridge Energy Detection (RED) algorithm. Score level fusion is used for combining the two modalities and Hamming Distance is applied for generating matching scores for both the traits. The combination of iris and palm print is a very powerful biometric trait due to the individual strengths and uniqueness of both the traits. The proposed work resulted in a Genuine Acceptance Rate (GAR) of 100% at a very low False Acceptance Rate (FAR) of 0.007 only. The Equal Error Rate (EER) value calculated is found to be 0.005. The values are obtained by testing the algorithm on three datasets i.e. Iris Image Dataset provided by IIT Delhi and two for palm print i.e. Palmprint database provided by COEP, touchless palmprint dataset provided by IIT Delhi.
The automatic recognition of an individual using the exclusive biometric trait like iris is the latest sensation in the field of biometric applications. The enormous mathematical advantages provided by iris pattern recognition make it hugely popular amongst the real time commercial applications which involve large scale databases. In recent years, we have witnessed several newly proposed methods attracting a lot of attention in the biometrics literature. Most of these methods basically use Daugman’s algorithm with a slight variation and claim to be very promising in terms of efficiency. However, there has been barely any effort to ensure the uniqueness of iris pattern and account for the rotational inconsistencies present in the iris images while capturing them. After analyzing the recently developed frameworks, the authors of this paper have proposed an unaccustomed approach which considers these aspects which are usually neglected in the methods proposed so far. This paper gives an insight of our system designed by a novel approach and presents the performance achieved by it over two standard iris image datasets of IIT Delhi and CASIA1. Our approach outperforms the well-known existing methods over these two different datasets on the basis of experimental results obtained. Moreover, it is illustrated that we have significantly reduced both the FAR and FRR at a unique separation point which is a remarkable accolade in this field. We have almost achieved perfect recognition with FAR and FRR of 0.001% and 0.069% respectively over the IIT Delhi and CASIA1 dataset.
Algorithms developed by the author for recognizing persons by their iris patterns have now been tested in many field and laboratory trials, producing no false matches in several million comparison tests. The recognition principle is the failure of a test of statistical independence on iris phase structure encoded by multi-scale quadrature wavelets. The combinatorial complexity of this phase information across different persons spans about 249 degrees of freedom and generates a discrimination entropy of about 3.2 b/mm<sup>2</sup> over the iris, enabling real-time decisions about personal identity with extremely high confidence. The high confidence levels are important because they allow very large databases to be searched exhaustively (one-to-many "identification mode") without making false matches, despite so many chances. Biometrics that lack this property can only survive one-to-one ("verification") or few comparisons. The paper explains the iris recognition algorithms and presents results of 9.1 million comparisons among eye images from trials in Britain, the USA, Japan, and Korea.
Improving Iris Recognition Performance using Acquisition, Segmentation and
- P V Dable
- P R Lakhe
- S S Kemekar
Iris Recognition Using Fusion of Gray Level Co-Occurrence Matrix and Gray Level Run Length Matrix with Svm and Knn Classifier
- luhadiya
Iris Authentication System using Hybrid
- plaka
Iris Segmentation Along with Noise Detection using Hough Transform
- S Singh
- Sh
- Singh
Iris Segmentation Along with Noise Detection using Hough Transform
- singh
Improving Iris Recognition Performance using Acquisition, Segmentation and
- dable