Delores M. Etter

Iris pattern extraction using bit planes and standard deviations

Iris recognition has been shown to be very accurate for human identification. In this paper, we develop a technique for iris pattern extraction utilizing the least significant bit-plane: the least significant bit of every pixel in the image. Through binary morphology applied to the bit-plane, the pupillary boundary of the iris is determined. The limbic boundary is identified by evaluating the standard deviation of the image intensity along the vertical and horizontal axes. Because our extraction approach restricts localization techniques to evaluating only bit-planes and standard deviations, iris pattern extraction is not dependent on circular edge detection. This allows for an expanded functionality of iris identification technology by no longer requiring a frontal view, which leads to the potential for off-angle iris recognition technology. Initial results show that it is possible to fit a close elliptical approximation to an iris pattern by using only bit-planes and standard deviations for iris localization.

Experimental results with increased bandwidth analysis filters in oversampled, subband acoustic echo cancelers

The motivation for adaptive filtering in subbands stems from two well-known problems in least-mean square full-band adaptive filtering. First, the convergence and tracking can be very slow if the input correlation matrix is ill conditioned, as in the case with speech input. Second, very high order adaptive filters are computationally expensive. One problem with adaptive filtering in subbands is the slow, asymptotic convergence associated with oversampled systems. Increasing the bandwidth of analysis filters relative to the synthesis filters is proposed to reduce the slow asymptotic convergence. The authors present experimental results illustrating the benefits of this modification.<<ETX>>

Analysis of partial iris recognition using a 1D approach

Iris recognition has been shown to be very accurate for human identification. We investigate the performance of the use of a partial iris for recognition. A partial iris identification system based on a one-dimensional approach to iris identification is developed. Experiment results show that a more distinguishable and individually unique signal is found in the inner rings of the iris. The results also show that it is possible to use only a portion of the iris for human identification.

Use of one-dimensional iris signatures to rank iris pattern similarities

A one-dimensional approach to iris recognition is presented. It is translation-, rotation-, illumination-, and scale-invariant. Traditional iris recognition systems typically use a two-dimensional iris signature that requires circular rotation for pattern matching. The new approach uses the Du measure as a matching mechanism, and generates a set of the most probable matches (ranks) instead of only the best match. Since the method generates one-dimensional signatures that are rotation-invariant, the system could work with eyes that are tilted. Moreover, the system will work with less of the iris than commercial systems, and thus could enable partial-iris recognition. In addition, this system is more tolerant of noise. Finally, this method is simple to implement, and its computational complexity is relatively low.

A new approach to iris pattern recognition

An iris identification algorithm is proposed based on adaptive thresholding. The iris images are processed fully in the spatial domain using the distinct features (patterns) of the iris. A simple adaptive thresholding method is used to segment these patterns from the rest of an iris image. This method could possibly be utilized for partial iris recognition since it relaxes the requirement of using a majority of the iris to produce an iris template to compare with the database. In addition, the simple thresholding scheme can improve the computational efficiency of the algorithm. Preliminary results have shown that the method is very effective. However, further testing and improvements are envisioned.

Iris recognition using histogram analysis

Iris recognition is perhaps the most accurate means of personnel identification due to the uniqueness of the patterns contained in each iris. Most commercial iris recognition systems use a patented algorithm based on two-dimensional Gabor wavelets developed by Daugman. This paper describes an alternate means to identify individuals using images of their iris. Here, we simplify the process by using preprocessed one-dimensional histograms. The methodology in forming these histograms, how they are used in enrollment and identification and performance in terms of false positives and false negatives are presented.

IIR algorithms for adaptive line enhancement

In this paper we introduce a simple IIR structure for the adaptive line enhancer. Two algorithms based on gradient-search techniques are presented for adapting the structure. Results from experiments which utilized real data as well as computer simulations are provided.

A multidisciplinary approach to biometrics

Biometrics is an emerging field of technology using unique and measurable physical, biological, or behavioral characteristics that can be processed to identify a person. It is a multidisciplinary subject that integrates engineering, statistics, mathematics, computing, psychology, and policy. The need for biometrics can be found in governments, in the military, and in commercial applications. The Electrical Engineering Department at the U.S. Naval Academy, Annapolis, MD, has introduced a biometric signal processing course for senior-level undergraduate students and has developed a biometrics lab to support this course. In this paper, the authors present the course content, the newly developed biometric signal processing lab, and the interactive learning process of the biometric course. They discuss some of the challenges that were encountered in implementing the course and how they were overcome. They also provide some feedback from the course assessment.

Wavelet basis reconstruction of nonuniformly sampled data

A well-documented problem in the analysis of real-world measurements is that the data may suffer from several problems, including data dropouts, an irregularly spaced sampling grid, and time-delayed sampling. These data irregularities render many traditional signal processing techniques unusable, and thus the data must be interpolated onto an even grid before scientific analysis techniques can be applied. We propose a method to perform a reconstruction of nonuniformly sampled signals using a wavelet basis fit in a multiresolutional setting. The advantage of using a wavelet basis is that we are able to not only reconstruct the signal using global information, but we are also able to take advantage of locally dense areas of sampling to reconstruct at higher resolutions.

A one-dimensional approach for iris identification

A novel approach to iris recognition is proposed in this paper. It differs from traditional iris recognition systems in that it generates a one-dimensional iris signature that is translation, rotation, illumination and scale invariant. The Du Measurement was used as a matching mechanism, and this approach generates the most probable matches instead of only the best match. The merit of this method is that it allows users to enroll with or to identify poor quality iris images that would be rejected by other methods. In this way, the users could potentially identify an iris image by another level of analysis. Another merit of this approach is that this method could potentially improve iris identification efficiency. In our approach, the system only needs to store a one-dimensional signal, and in the matching process, no circular rotation is needed. This means that the matching speed could be much faster.