@rnsit.ac.in
Principal
R. N. Shetty Institute of Technology (RNSIT)
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
G N Srikanth and M K Venkatesha
Universidad Tecnica de Manabi
Speech has information more than text, but under noisy environment speech sufferance from disadvantage of not properly decoded by humans and same is true with machines. speech being bimodal along with audio features if we augment visual features specifically related to lip movements. the degree of speech recognition can be improved. The objective of this work is to use audio and visual features to aid word recognition. In this work we extracted MFCC features for audio and Geometrical features of lip movements together is used in machine learning algorithm to predict the word utterances. Videos related to word utterances are extracted from TIMID database. With the statistical information related to audio and corresponding visual features from lip movements is extracted to form input feature vector to machine learning algorithm (Multi-layer perceptron). The experimental results show that using MLP we have obtained a word recognition accuracy of 91% and using KNN Classifier the accuracy attained is 61%. The results presented here have important implications for applications in HMI communication and helps hearing impaired.
M. J. Sudhamani, Ipsita Sanyal, and M. K. Venkatesha
Springer Singapore
C. Anitha, M.K. Venkatesha, and B. Suryanarayana Adiga
IEEE
For any real-time application, detection and tracking of features becomes very important. The detection and tracking algorithms have to be very robust and efficient with least or zero false positives and false negatives. We use a novel combination for detection and tracking purpose. In this paper we propose a robust mouth region extraction and tracking algorithm that works in real-time. The region of interest for our application requires the face and mouth regions. We propose a novel technique for extracting the mouth region automatically. The proposed technique detects and tracks the mouth in either closed or opened state. We use the color components for skin tone and lips extraction.
C. Anitha, M.K. Venkatesha, and B. Suryanarayana Adiga
Elsevier BV
Abstract One of the prominent indicators of drowsiness is yawning. The main intention for a real-time application such as detecting the driver's yawning is that the response of the detector must be as quick as possible. A novel yawning detection system is proposed which is based on a two agent expert system. The features of the face have to be extracted to detect yawning in the driver's face. In the proposed system, as the first part of detection we use the face detection algorithm's skin detection part. The skin region is extracted. For all the skin region blocks detected, their boundaries are defined. Then segmented face is divided into two halves. The lower half of the face is considered for the mouth region extraction. The presence of yawning would be indicated by a black blob in the mouth region of the binary image. But, there may be multiple blobs present in the image which may be due to the presence non-skin like regions around the driver's face. So, identifying the exact position of the mouth and checking for its containment inside the face is necessary. The features extracted for yawning detection are the histogram values taken from the vertical projection of the lower part of the face.
Sudhamani M J, M K Venkatesha, and Radhika K R
IEEE
A Multimodal authentication system with fusion at decision level employing Iris and Finger Vein is developed. This paper also proposes a novel feature extraction technique for finger vein and iris traits. This approach combines the decisions from the individual modalities using the conventional AND rule. Experimental results demonstrate the effectiveness of the proposed multimodal authentication system for accurate person authentication.
H. R. Bhagyalakshmi and M. K. Venkatesha
Springer India
Reversible logic gates are in high demand in the world of low-power digital circuits and quantum computers since the power dissipation problems which occur in the computational operations of a computer can be avoided by using reversible structures instead of the conventional irreversible logic structures. One important task is to compare the inputs using a computer and decide what to do next. In a quantum computer, this must be built using reversible logic circuits. In this paper, we are presenting a two-bit comparator using new reversible logic gate called BVMF gate which is a multifunction gate. The optimization of the designed circuit is achieved by keeping the number of gates and number of garbage outputs to a minimum value.
M J Sudhamani, M K Venkatesha, and K R Radhika
IEEE
The primary application of biometric technology is to analyze human characteristics for security purpose as security issues have been the focused interest in recent years. A survey of different fusion techniques in multimodal biometrics is made. This paper also attempts to identify some of the challenges and issues that confront research in multimodal biometrics.
K.R. Radhika, M.K. Venkatesha, and G.N. Sekhar
Elsevier BV
Abstract: Current network based authentication applications require simple robust methods which are faster and does not choke the bandwidth. Minimal features are considered using subpattern analysis which leads to less response time in a real time scenario. Certain subsections of signature vary in a genuine case subdivision process. With a high degree of certainty the minimum variance quadtree components [MVQCs] of a signature for a person are listed to apply on testing sample. Hu moments are applied on the selected subsections. The summation values of the subsections are provided as feature to radial basis function [RBF] and feed forward neural network classifiers. Results indicate that the Radial Basis Function classifier yielded 7% false rejection rate and feed forward neural network classification technique produced 9% false rejection rate. The major advantage of proposed system is, storage of person dependent template details are drastically reduced. Only the MVQC list of a person is to be cataloged.
K.R. Radhika, M.K. Venkatesha, and G.N. Sekhar
Elsevier BV
In this work, shape analysis of the acceleration plot, using lower order Zernike moments is performed for authentication of on-line signature. The on-line signature uses time functions of the signing process. The lower order Zernike moments represent the global shape of a pattern. The derived feature, acceleration vector is computed for the sample signature which comprises on-line pixels. The Zernike moment represent the shape of the acceleration plot. The summation value of a Zernike moment for a signature sample is obtained on normalized acceleration values. This type of substantiation decreases the influence of primary features with respect to translation, scaling and rotation at preprocessing stage. Zernike moments provide rotation invariance. In this investigation it was evident that the summation of magnitude of a Zernike moment for a genuine sample was less as compared to the summation of magnitude of a imposter sample. The number of derivatives of acceleration feature depends on the structural complexity of the signature sample. The computation of best order by polynomial fitting and reference template of a subject is discussed. The higher order derivatives of acceleration feature are considered. Signatures with higher order polynomial fitting and complex structure require higher order derivatives of acceleration. Each derivative better represents a portion of signature. The best result obtained is 4% of False Rejection Rate [FRR] and 2% of False Acceptance Rate [FAR].
K. R. RADHIKA, M. K. VENKATESHA, and G. N. SEKHAR
World Scientific Pub Co Pte Lt
This paper proposes an online signature authentication based on polynomial modeling. The derived feature acceleration is used. The order of the polynomial is carefully selected for each subject using the acceleration vector. The dynamic warping algorithm and mean squared error measure aid in standard template selection process for a subject. The best fit order is directly dependent on the cursive information of the signature sample. We provide a practical analysis of residual term, number of zeros inside the unit circle and distance measure on pole values for the purpose of classification. Storing order and standard template for a subject, reduces the runtime and memory requirements for a dynamic system. The best performance was achieved based on the number of zeros inside unit circle with false rejection rate as 5% and false acceptance rate as 7%. The aim of the work is to provide simple robust authentication suitable for hand held devices. Even compared to state-of art experiments, the proposed system is aimed to provide authentication using a single derived feature.
K. R. RADHIKA, S. V. SHEELA, M. K. VENKATESHA, and G. N. SEKHAR
World Scientific Pub Co Pte Lt
Authentication systems which are covenant with a measurable behavioral and physiological traits are essential for an online system. In this paper, two types of biometric sample authentication from different databases on a common algorithm using Continuous Dynamic Programming [CDP] are discussed. Using a common algorithm, a method for user-dependent threshold decisions can be achieved for both biometrics in a uniform fashion. The integration of static iris information and dynamic signature information are done at decision level. Inferences are drawn using voting techniques. The derived kinematic feature, acceleration, is used in this paper.
Radhika
Science Publications
Problem statement: The research addressed the computational load reduction in off-line signature verification based on minimal features using bayes classifier, fast Fourier transform, linear discriminant analysis, principal component analysis and support vector machine approaches. Approach: The variation of signature in genuine cases is studied extensively, to predict the set of quad tree components in a genuine sample for one person with minimum variance criteria. Using training samples, with a high degree of certainty the Minimum Variance Quad tree Components (MVQC) of a signature for a person are listed to apply on imposter sample. First, Hu moment is applied on the selected subsections. The summation values of the subsections are provided as feature to classifiers. Results: Results showed that the SVM classifier yielded the most promising 8% False Rejection Rate (FRR) and 10% False Acceptance Rate (FAR). The signature is a biometric, where variations in a genuine case, is a natural expectation. In the genuine signature, certain parts of signature vary from one instance to another. Conclusion: The proposed system aimed to provide simple, faster robust system using less number of features when compared to state of art works.
K R Radhika, M K Venkatesha, and G N Shekar
IEEE
Zernike moments are image descriptors often used in pattern recognition. They offer rotation invariance. In this paper, we discuss a novel method of signature authentication using Zernike moments. Instead of working on primary features such as image or on-line data, working on the derived kinematic plot is a robust way of authentication. The derived kinematic plot considered in this paper is acceleration plot. Each signature's on-line acceleration information is being weighted by Zernike moment. The shape analysis of the acceleration plot, using only lower order Zernike moments is performed for authentication of on-line signature.
K. R. Radhika, S. V. Sheela, M. K. Venkatesha, and G. N. Sekhar
Springer Berlin Heidelberg
Storing and retrieving the behavioral and physiological templates of a person for authentication using a common algorithm is indispensable in on-line applications. This paper deals with authentication of on-line signature data and textual iris information using continuous dynamic programming [CDP]. Kinematic derived feature, acceleration is considered. The shape of acceleration plot is analysed. The experimental study depict that, as the number of training samples considered for CDP algorithm increase, the false rejection rate decrease.
P. V. Manjunath, H. R. Baghyalakshmi, and M. K. Venkatesha
IEEE
A new low power low voltage CMOS thyristor delay element is proposed in this paper. The delay range of the proposed circuit is extended by reducing charge sharing problem. The delay element has less voltage and temperature sensitivity and consumes less power. The circuit is implemented in 130nm technology and simulation result shows that the delay range is from 180ps to 9ns which is very high compared to other architectures of CMOS thyristor delay element. The power consumption is in the range of 24µW to 243µW. This circuit also has low voltage and temperature sensitivity compared to previous circuits. A digitally controlled current source is implemented to generate control current in the range of 1µA to 32µA. A cascode current mirror is used to couple the control current to the delay element circuit.
S. V. Sheela, K. R. Radhika, M. K. Venkatesha, and P. A. Vijaya
IEEE
In online applications, authentication systems which are covenant with a measurable behavioral trait and physiological characteristics are essential. This paper deals with authenti- cation of an individual's on-line signature data and textual iris information using continuous dynamic programming (CDP). Instead of working on primary features such as image or on-line data, working on the derived kinematic plot is robust way of authentication. I. INTRODUCTION Signature verification and Iris recognition, as biometric technologies have great advantages such as variability, scal- ability and security thus providing a variety of applications. Secure communications for mortgage, passport, internet com- merce and mobile commerce are some of the areas which thrive on recognition for payments, logins via a tablet PC, crypto-biometrics and for bio-hashing. New recognition tech- niques that can mine and discover behavioral knowledge in large data sets are very much essential. Challenging concept of signature authentication is that, it is strongly affected by user-dependencies as it varies from one signing instance to another in a known way. The iris recognition system in an unconstrained environment due to quality of pictures, bad lighting, and occlusion by eyelids, noises and inappropriate eye positioning is still far from perfection. Research issues are based on iris localization, nonlinear normalization, occlusion segmentation, liveness detection, large scale identification. Some of the various approaches of on-line signature verifi- cation systems as reported in literature are broadly divided into four classes as global parametric feature based approach, function-based approach, hybrid methods based on both of the above stated schemes and trajectory construction methods. Feature-based approaches are statistical parametric methods in which a holistic vector representation consisting of a set of global features is derived from the signature trajectories. Function-based approaches are methods in which time se- quences describing local properties of the signature are used for recognition. Trajectory construction methods produce and control complex two-dimensional synergistic movements. Iris recognition methods are classified into phase-based method, texture-analysis based method, zero-crossing representation method, approach based on local intensity variations and approach using Independent Component Analysis(ICA). In phase-based method the iris pattern is demodulated to extract phase information using quadrature 2D Gabor wavelets (1). Texture-analysis based method uses Laplacian of Gaussian applied to the image at multiple scales (2). The zero-crossing representation method deals with the representation of features of the iris at different resolution levels based on the wavelet transform zero crossing representation (3). In the approach based on local intensity variations the sharp variation points of iris patterns are recorded as features and are utilized to represent the characteristics of the iris (4). In the approach using ICA the independent components are uncorrelated and the feature coefficients are considered to be nongaussian and mutually independent (5).
K. R. Radhika, M. K. Venkatesha, and G. N. Sekhar
IEEE
In online applications, authentication systems which are covenant with a measurable behavioral trait is essential. The flow of signature with respect to time is termed as on- line signature data. This paper deals with authentication of an individual's on-line signature using continuous dynamic pro- gramming (CDP). Modern systems aim to move security from simple static passwords to more dynamic security measures to suit the comfort level of the user in mobile commerce and web commerce. Recognition of individual's signature is text dependent self-certification process with constant behavioral variation. I. INTRODUCTION
K. R. Radhika, G. N. Sekhar, and M.K. Venkatesha
IEEE
Over the last two decades, in the field of secure communication and financial on-line applications, we have witnessed an explosive growth in biometric personal authentication systems which are covenant with a measurable physical characteristic or behavioral trait. On-line refers to making use of the time functions of the signing process. Verification of on-line signature as a biometric modality still is challenging fields of research, since the number of pointer based devices emerges as input devices for many e-commerce and m-commerce applications. The deployment of automatic handwritten signature verification with technology still remains open for novel methods due to inter-class and intra-class variations of signature. Since the area is currently one of the most on the go and the bulk of research is very large, this survey paper covers some of the examples of the ways.
K. R. Radhika, G. N. Sekhar, and M. K. Venkatesha
IEEE
In online applications, there is a foreseeable explosive growth in biometric personal authentication systems which are covenant with a measurable behavioral trait. Hand written object [HO] verification is the process used to recognize an individual, which is intuitive reliable indicator. Verification of a HO as a biometric modality still is a challenging field of research, as number of online and offline commercial applications use modern acquisition devices. The employment of HO verification with technology still remains open for novel methods due to inter-class and intra-class variations. This paper discusses the trajectory generation [TG] methods applicable for any HO verification, such as character recognition, handwriting verification, style classification, shape recognition and signature recognition, which are suitable for latest trend of mobile-commerce and web-commerce applications.