Ritu Vijay

Image Quality Prediction by Minimum Entropy Calculation for Various Filter Banks

implement image compression using various wavelet filter banks and measure performance with rate distortion characterizations. Various separable filter banks are chosen and compared. Coefficients in the subbands obtained by wavelet decomposition are quantized. The image is then reconstructed from the quantized coefficients, and distortion is measured. Three distortion measures are used: Entropy of reconstructed image, energy retained (ER) and redundancy.

Online acquisition and wireless transmission of carotid pulse waveforms to analyse posture related changes

The physiological parameters of the fighter plane pilot undergo drastic changes leading to G stress accumulation. Heart rate is considered primary mechanism which plays a great role in acceleration protection research especially for G-LOC (G-induced loss of consciousness). Various methods based on various principles have been developed to study heart rate so far. Carotid pulse waveforms obtained from the human subject with the help of piezoelectric transducer may be used to analyse the cardiac activity. Change in heart rate due to change in the posture may be used to explain G stress effect on humans. Carotid signal is acquired by placing piezoelectric sensor based system on the subject’s carotid artery. The piezoelectric sensor has a transducing element which is efficient and effective to sense movement due to carotid pulse and produce piezoelectric signal data in response thereto. The real time data of carotid pulse waveforms of 12 human subjects are acquired and filtered using MATLAB software in various body postures. RR period and pulse amplitudes are quantified from real time recorded carotid signal by visual inspection. Further Statistical analysis is done to calculate the heart rate. The filtered stored data is transmitted wirelessly from one PC to another PC using Teamviewer software.

Rank-level Fusion of Multispectral Palmprints

This paper presents an approach for the personal authentication using rank-level fusion of multispectral palmprints, instead of using multiple biometric modalities and multiple matchers. The rank level fusion involving the non linear combination of hyperbolic tangent functions gives the best recognition rate for the Rank 1 obtained from two types of features, viz., sigmoid and fuzzy. The results of using rank level fusion on the publicly available multispectral palmprint database show the significant improvement in the recognition rate as compared to the individual spectral bands. Recognition rate of 99.4% from sigmoid features and that of 99.2% from fuzzy features based on Rank 1 is the outcome of the hyperbolic tangent nonlinearity.

Analysis of Orthogonal and Biorthogonal Wavelet Filters for Image Compression

e present work we analyze the performance of orthogonal and Biorthogonal wavelet filters for image compression on variety of test images. The test images are of different size and resolution. The compression performance is measured, objectively peak signal to noise ratio and subjectively visual quality of image and it is found that Biorthogonal wavelets outperform the orthogonal ones in both the criteria. General Terms Image compression algorithms

A finger-knuckle-print authentication system based on DAISY descriptor

This paper presents an approach for finger knuckle print (FKPs) authentication using DAISY descriptor on different knuckle combinations. Score level fusion is applied on scores of features extracted from FKPs. An improved identification rate is achieved after fusion as compared to that from the individual knuckle print. There has also been a little effort on study of feature selection. The entropy measures are employed for the feature selection. The experimental results also demonstrate the superiority of the DAISY descriptor over other features in terms of the time and accuracy.

Rough Set Approach in Machine Learning: A Review

The Rough Set (RS) theory can be considered as a tool to reduce the input dimensionality and to deal with vagueness and uncertainty in datasets. Over the years, there has been a rapid growth in interest in rough set theory and its applications in artificial intelligence and cognitive sciences, especially in research areas such as machine learning, intelligent systems, inductive reasoning, pattern recognition, data preprocessing, knowledge discovery, decision analysis, and expert systems. This paper discusses the basic concepts of rough set theory and point out some rough set-based research directions and applications. The discussion also includes a review of rough set theory in various machine learning techniques like clustering, feature selection and rule induction. General Terms Information and decision systems, (in)discernibility , approximation spaces, rough sets, rough membership functions, reducts, decision rules, dependencies of attributes. Clustering, Rule Induction, Feature Selection

Acoustic analysis of speech under stress

When a person is emotionally charged, stress could be discerned in his voice. This paper presents a simplified and a non-invasive approach to detect psycho-physiological stress by monitoring the acoustic modifications during a stressful conversation. Voice database consists of audio clips from eight different popular FM broadcasts wherein the host of the show vexes the subjects who are otherwise unaware of the charade. The audio clips are obtained from real-life stressful conversations (no simulated emotions). Analysis is done using PRAAT software to evaluate mean fundamental frequency (F0) and formant frequencies (F1, F2, F3, F4) both in neutral and stressed state. Results suggest that F0 increases with stress; however, formant frequency decreases with stress. Comparison of Fourier and chirp spectra of short vowel segment shows that for relaxed speech, the two spectra are similar; however, for stressed speech, they differ in the high frequency range due to increased pitch modulation.

Hamming Distance based Clustering Algorithm

Cluster analysis has been extensively used in machine learning and data mining to discover distribution patterns in the data. Clustering algorithms are generally based on a distance metric in order to partition the data into small groups such that data instances in the same group are more similar than the instances belonging to different groups. In this paper the authors have extended the concept of hamming distance for categorical data.As a data processing step they have transformed the data into binary representation. The authors have used proposed algorithm to group data points into clusters. The experiments are carried out on the data sets from UCI machine learning repository to analyze the performance study. They conclude by stating that this proposed algorithm shows promising result and can be extended to handle numeric as well as mixed data.

A Comprehensive Study of Palmprint based Authentication

This paper presents some new features for the palmprint based authentication. The Region of interest (ROI) is extracted from the palmprint image by finding a tangent to the curves between fingers. The perpendicular bisector of this tangent and the tangent itself help demarcate the rectangular area that forms the ROI of the palmprint. Four approaches are presented for the feature extraction. In the first approach the ROI is divided into a suitable number of non-overlapping windows from which fuzzy features are extracted. In the second approach multi-scale wavelet decomposition is applied on the ROI and the detail images are combined to yield a composite image which is partitioned into non-overlapping windows and energy features are extracted. In the third approach sigmoid features are extracted from the ROI and in the fourth approach feature extraction is done using Local Binary Pattern (LBP) based on the directional gradient response. These four sets of features are used for the authentication of users from two databases using Euclidean Distance, Chi square measure and Support Vector Machines as classifiers.

A Bi-objective Genetic Algorithm Optimization of Chaos-DNA Based Hybrid Approach

Abstract The paper implements and optimizes the performance of a currently proposed chaos-deoxyribonucleic acid (DNA)-based hybrid approach to encrypt images using a bi-objective genetic algorithm (GA) optimization. Image encryption is a multi-objective problem. Optimizing the same using one fitness function may not be a good choice, as it can result in different outcomes concerning other fitness functions. The proposed work initially encrypts the given image using chaotic function and DNA masks. Further, GA uses two fitness functions – entropy with correlation coefficient (CC), entropy with unified average changing intensity (UACI), and entropy with number of pixel change rate (NPCR) – simultaneously to optimize the encrypted data in the second stage. The bi-objective optimization using entropy with CC shows significant performance gain over the single-objective GA optimization for image encryption.