Selva Kumari

Learn More
This paper motivates the use of Dynamic Mel-Frequency Cepstral Coefficient (DMFCC) feature and combination of DMFCC and MFCC features for robust language and text-independent speaker identification. MFCC feature, modeled on the human auditory system has been the widely used feature for speaker recognition because of its less vulnerability to noise(More)
  • R Shantha, Selva Kumari, A Thilagamanimala, P G Student, V Sadasivam
  • 2007
Wavelet-based signal processing has become commonplace in the signal processing community over the past decade. One of the most important applications of wavelets is removal of noise from signals called denoising accomplished by thresholding wavelet coefficients in order to separate signal from noise. However, conventional denoising fails for signals with(More)
Pseudomonas syringae pv. tomato DC3000 (PtoDC3000) is an extracellular model plant pathogen, yet its potential to produce secreted effectors that manipulate the apoplast has been under investigated. Here we identified 131 candidate small, secreted, non-annotated proteins from the PtoDC3000 genome, most of which are common to Pseudomonas species and(More)
A novel method for unsupervised change detection in multi-temporal satellite images using Gaussian mixture model (GMM) with spatial information is proposed. This approach is based on three steps. Firstly, the difference image between two Synthetic Aperture Radar (SAR) images of the same area taken at two different times is obtained using the standard(More)
  • R Meena Prakash, R Shantha, Selva Kumari, M Vimala, Meena Prakash
  • 2015
In this paper, a new method for segmentation of images based on Fuzzy C Means Clustering is presented. The fuzzy c means is one of the clustering based methodologies. It has been extensively used for segmentation of images. Compared with the Hard C means clustering algorithm, it can retain more information in the image because of introducing fuzziness. The(More)
The Segmentation of the Images refers to extracting the needed region from the image based on some specified methodologies. Thresholding Approach, Model-based Approach, Level Set Approach are some of the segmentation methodologies. The clustering methodologies can provide accurate results for most of the cases. As the number of clusters separated from the(More)
  • S Selva Nidhyananthan, M Pradeepa, V Mathangi, Mathi Vathana, R Shantha, Selva Kumari
  • 2015
A new approach for joint multichannel speech source separation is proposed in this paper. The technique starts by creating a database in the context of multiple simultaneous speakers in the reverberant overcomplete environment. Recordings are made in overcomplete environment where number of sources are greater than number of observations. The speech of(More)
  • Corresponding Author, Sundaradhas, Selva Nidhyananthan, Sundaradhas Selva, Nidhyananthan, Ramapackiam Shantha +1 other
  • 2014
This article evaluates the performance of Extreme Learning Machine (ELM) and Gaussian Mixture Model (GMM) in the context of text independent Multi lingual speaker identification for recorded and synthesized speeches. The type and number of filters in the filter bank, number of samples in each frame of the speech signal and fusion of model scores play a(More)
The trends toward low voltage low power silicon chip system have been growing due to the increasing demand of smaller size and long battery life for biomedical applications. This paper deals with well-defined design criteria for CMOS current mode operational amplifiers. A simple design procedure is presented which has current mode preamplifier with an(More)
Computers' required random numbers initially, for simulations and numerical computations like Monte Carlo calculations. Random number generators offer an important contribution to many communication systems for security. They are critical components in computational science. However the tradeoff between quality and computational performance is an issue for(More)