Raghunath S. Holambe

Half-Iris Feature Extraction and Recognition Using a New Class of Biorthogonal Triplet Half-Band Filter Bank and Flexible k-out-of-n:A Postclassifier

Abstract-This paper presents a shift, scale, and rotation-in- variant technique for iris feature-representation and fused postclassification at the decision-level to improve the accuracy and speed of the iris-recognition system. Most of the iris-recognition systems are still incapable for providing low false rejections due to a wide variety of artifacts and are computationally inefficient. In order to address these problems, effective and computationally efficient iris features are extracted based on a new class of triplet half-band filter bank (THFB). First, a new class of THFB is designed by using generalized half-band polynomial suitable for iris feature extraction. This THFB satisfies perfect reconstruction (PR) and provides linear phase, regularity, better frequency-selectivity, near-orthogonality, and good time-frequency localization. The uses of these properties are investigated to approximate iris features significantly. Second, a novel flexible k-out-of-n.A (Accept) postclassifier (any k-out-of-n-regions-Accept) is explored to achieve the robustness against possible intraclass iris variations. The proposed approach (THFB+ k-out-of-n.A) is capable of handling various artifacts, particularly segmentation error, eyelid/eyelashes occlusion, shadow of eyelids, head-tilt, and specular reflections during iris verification. Experimental results using UBIRIS, MMU1, CASIA-IrisV3, and IITD databases show the superiority of the proposed approach with some of the existing popular iris-recognition algorithms.

Rapid and brief communication: Orientation feature for fingerprint matching

This paper describes a fingerprint verification algorithm based on the orientation field. The orientation field of a fingerprint image has also been used for image alignment. Area around the core point has been employed as an area of interest for determining the orientation feature map. The algorithm has been tested on two databases (database available from University of Bologna, Biometrics Laboratory and FVC2002). The performance of the algorithm is measured in terms of receiver operating characteristics (ROC). For the University of Bologna database, at ~0% false acceptance rate (FAR) the genuine acceptance rate (GAR) observed is ~78% and at ~11% FAR, GAR is ~97%. For the FVC2002 database at ~0% FAR the GAR observed is 75% and at ~18% FAR, GAR is 93%. Proposed algorithm yields better GAR at low FAR with reduced computational complexity. Because of simplicity in computations the algorithm can be easily implemented as an embedded automatic fingerprint identification system (AFIS).

DWT and LPC based feature extraction methods for isolated word recognition

In this article, new feature extraction methods, which utilize wavelet decomposition and reduced order linear predictive coding (LPC) coefficients, have been proposed for speech recognition. The coefficients have been derived from the speech frames decomposed using discrete wavelet transform. LPC coefficients derived from subband decomposition (abbreviated as WLPC) of speech frame provide better representation than modeling the frame directly. The WLPC coefficients have been further normalized in cepstrum domain to get new set of features denoted as wavelet subband cepstral mean normalized features. The proposed approaches provide effective (better recognition rate), efficient (reduced feature vector dimension), and noise robust features. The performance of these techniques have been evaluated on the TI-46 isolated word database and own created Marathi digits database in a white noise environment using the continuous density hidden Markov model. The experimental results also show the superiority of the proposed techniques over the conventional methods like linear predictive cepstral coefficients, Mel-frequency cepstral coefficients, spectral subtraction, and cepstral mean normalization in presence of additive white Gaussian noise.

Text-independent speaker identification using Radon and discrete cosine transforms based features from speech spectrogram

This paper presents a new feature extraction technique for speaker recognition using Radon transform (RT) and discrete cosine transform (DCT). The spectrogram is compact, efficient in representation and carries information about acoustic features in the form of pattern. In the proposed method, speaker specific features have been extracted by applying image processing techniques to the pattern available in the spectrogram. Radon transform has been used to derive the effective acoustic features from the speech spectrogram. Radon transform adds up the pixel values in the given image along a straight line in a particular direction and at a specific displacement. The proposed technique computes Radon projections for seven orientations and captures the acoustic characteristics of the spectrogram. DCT applied on Radon projections yields low dimensional feature vector. The technique is computationally efficient, text-independent, robust to session variations and insensitive to additive noise. The performance of the proposed algorithm has been evaluated using the Texas Instruments and Massachusetts Institute of Technology (TIMIT) and our own created Shri Guru Gobind Singhji (SGGS) databases. The recognition rate of the proposed algorithm on TIMIT database (consisting of 630 speakers) is 96.69% and for SGGS database (consisting of 151 speakers) is 98.41%. These results highlight the superiority of the proposed method over some of the existing algorithms.

A Blind DCT Domain Digital Watermarking for Biometric Authentication

In this paper, an efficient blind digital image watermarking algorithm using mapping technique is presented. The algorithm can embed or hide an entire image or pattern (logo) directly into the original image. The embedding process is based on changing the selected DCT coefficients of the host image to odd or even values depending on the binary bit value of watermark DCT coefficients. The algorithm is tested for fingerprint image embedded with a face watermark. It is demonstrated that the watermarking algorithm offers a significant advantage of providing biometric image compression and authentication without introducing any significant degradation in the image quality. Moreover the watermarking scheme is blind and does not require any additional data for logo extraction.

Design of Low-Complexity High-Performance Wavelet Filters for Image Analysis

This paper addresses the construction of a family of wavelets based on halfband polynomials. An algorithm is proposed that ensures maximum zeros at for a desired length of analysis and synthesis filters. We start with the coefficients of the polynomial and then use a generalized matrix formulation method to construct the filter halfband polynomial. The designed wavelets are efficient and give acceptable levels of peak signal-to-noise ratio when used for image compression. Furthermore, these wavelets give satisfactory recognition rates when used for feature extraction. Simulation results show that the designed wavelets are effective and more efficient than the existing standard wavelets.

Text-Independent Speaker Identification Using Hidden Markov Models

This paper presents a closed-set, text-independent speaker identification using continuous density hidden Markov model (CDHMM). Each registered speaker has a separate HMM which is trained using Baum-Welch algorithm. The system performance has been studied for different system parameters such as the number of states, number of mixture components per state and the amount of data required for training. Identification accuracy of 100% is achieved by conducting the experiments on TIMIT database.

Fractional Fourier transform based features for speaker recognition using support vector machine

This paper presents a text-independent speaker recognition technique in which the conventional Fourier transform in Mel-Frequency Cepstral Coefficient (MFCC) front-end is substituted by fractional Fourier transform. Support Vector Machine (SVM) maps these input features into a high-dimensional space to separate classes by a hyperplane with enhanced discrimination capability. SVM based on mean-squared error classifier produces more accurate system. The Fractional Fourier Transform (FrFT) reveals the mixed time and frequency components of the signal. Modelling of speech signals as mixed time and frequency signals represents better production and perception speech characteristics. Processing of time-varying signals in fractional Fourier domain allows us to estimate the signal with least Mean Square Error (MSE) making the technique robust against additive noise compared to Fourier domain maintaining same computational complexity. The feasibility of the proposed technique has been tested experimentally using Texas Instruments and Massachusetts Institute of Technology (TIMIT) and Shri Guru Gobind Singhji (SGGS) databases. The experimental results show the superiority of the proposed method.

Partial iris feature extraction and recognition based on a new combined directional and rotated directional wavelet filter banks

This paper presents a novel approach to construct two-dimensional (2-D) non-separable, non-redundant, multiscale combined directional wavelet filterbank (CDWFB) for iris feature-extraction. This CDWFB is obtained by the combination of directional wavelet filterbank (DWFB) and rotated directional wavelet filterbank (RDWFB). Firstly, 2-D biorthogonal wavelet filterbank (BWFB) is designed based on the factorization of a general half-band polynomial. Secondly, McClellan transformation is used to obtain checkerboard shaped filterbank (CSFB) using designed BWFB coefficients. This CSFB is applied on 2-D BWFB to obtain DWFB. RDWFB is obtained using DWFB coefficients whose directions are 45^o apart from DWFB. Iris recognition systems are still incapable for providing low false rejection and significant representation. In order to address these problems, a novel approach is proposed to extract iris texture in twelve-directions by CDWFB. The inner half-iris region (partial iris) is divided into six non-overlapping sub-regions and selected four-regions for further processing to derive compact and significant iris-features. An independent feature extraction using CDWFB is carried out on each region. The dissimilarity measure of each region are fused at the decision level by exploring 1-out-of-n: Accept (A) post-classifier in order to reduce the false rejection rate. Experimental results using UBIRIS and MMU1 databases show the superiority of the proposed method with some of the popular iris recognition algorithms.

Speaker Identification Based on Robust AM-FM Features

Linear source-filter models have been widely used by researchers as a front-end for speaker identification systems. It uses the cepstral features derived from the power spectrum of the speech signal. But it is also well known that a significant part of the acoustic information cannot be modeled by the linear sourcefilter model, and thus, the need for nonlinear features becomes apparent. In this paper, an attempt is made to investigate the use of phase function in the analytic signal for deriving a representation of frequencies present in the speech signal. The main objective of the paper is to present a novel parameterization of speech that is based on the nonlinear AM-FM speaker model in the context of close-set speaker identification. The proposed features measure the amount of amplitude and frequency modulation and attempt to model aspects of the speaker related information that the commonly used linear source-filter model fails to capture. To evaluate the robustness of the proposed features for speaker identification, clean speech corpus from TIMIT database has been used and combined the speech signal with car noise and babble noise from the NOISEX-92 database. The proposed feature set provides significant improvements in the identification accuracy over the conventional method like MFCC under mismatched training and testing environments. The results show that better speaker identification rates are attainable under mismatched conditions especially at low signal-to-noise ratio (SNR).