Md. Khademul Islam Molla

Single-Mixture Audio Source Separation by Subspace Decomposition of Hilbert Spectrum

A novel technique is developed to separate the audio sources from a single mixture. The method is based on decomposing the Hilbert spectrum (HS) of the mixed signal into independent source subspaces. Hilbert transform combined with empirical mode decomposition (EMD) constitutes HS, which is a fine-resolution time-frequency representation of a nonstationary signal. The EMD represents any time-domain signal as the sum of a finite set of oscillatory components called intrinsic mode functions (IMFs). After computing the spectral projections between the mixed signal and the individual IMF components, the projection vectors are used to derive a set of spectral independent bases by applying principal component analysis (PCA) and independent component analysis (ICA). A k-means clustering algorithm based on Kulback-Leibler divergence (KLd) is introduced to group the independent basis vectors into the number of component sources inside the mixture. The HS of the mixed signal is projected onto the space spanned by each group of basis vectors yielding the independent source subspaces. The time-domain source signals are reconstructed by applying the inverse transformation. Experimental results show that the proposed algorithm performs separation of speech and interfering sound from a single mixture

Separation of EOG artifacts from EEG signals using bivariate EMD

A problem of eye-movement muscular interference removal from EEG recordings is described. In many experiments in neuroscience it is crucial to separate different sources of electrical activity within human body in a situation when a very limited knowledge about nonlinear and nonstationary nature of the mixing process is available. A new two step extension to bivariate empirical mode decomposition is proposed to remove ocular artifacts from EEG with a use of fractional Gaussian noise as a reference first to preprocess EOG signal, which is next used in the second step as a reference to clean EEG signals. Results with EEG experimental data validate the proposed approach.

Artifact suppression from EEG signals using data adaptive time domain filtering

This paper presents a data adaptive filtering approach to separate the electrooculograph (EOG) artifact from the recorded electroencephalograph (EEG) signal. Empirical mode decomposition (EMD) technique is used to implement the time domain filter. Fractional Gaussian noise (fGn) is used here as the reference signal to detect the distinguish feature of EOG signal to be used to separate from EEG. EMD is applied to the raw EEG and fGn separately to produce a finite number band limited signals named intrinsic mode functions (IMFs). The energies of individual IMFs of fGn and that of raw EEG are compared to derive the energy based threshold for the suppression of EOG effects. The separation results using EMD based approach is also compared with wavelet thresholding technique. The experimental results show that the data adaptive filtering technique performs better than the wavelet based approach.

Empirical mode decomposition analysis of climate changes with special reference to rainfall data

We have used empirical mode decomposition (EMD) method, which is especially well fitted for analyzing time-series data representing nonstationary and nonlinear processes. This method could decompose any time-varying data into a finite set of functions called “intrinsic mode functions” (IMFs). The EMD analysis successively extracts the IMFs with the highest local temporal frequencies in a recursive way. The extracted IMFs represent a set of successive low-pass spatial filters based entirely on the properties exhibited by the data. The IMFs are mutually orthogonal and more effective in isolating physical processes of various time scales. The results showed that most of the IMFs have normal distribution. Therefore, the energy density distribution of IMF samples satisfies χ2-distribution which is statistically significant. This study suggested that the recent global warming along with decadal climate variability contributes not only to the more extreme warm events, but also to more frequent, long lasting drought and flood.

Multivariate EMD based approach to EOG artifacts separation from EEG

Measured electroencephalography (EEG) signals can be contaminated with other electrophysiological signal sources. This contamination decreases accuracy of neuroengineering applications such as brain computer interfaces. This paper focuses on the removal of electrooculography (EOG) that strongly appears in frontal electrodes EEG. To develop an EOG removal algorithm, we propose to utilize recently developed a multivariate extension of empirical mode decomposition (EMD) called MEMD. MEMD decomposes a multichannel signal into a set of intrinsic mode functions (IMF), and the number of IMFs is identical among the channels. We establish a criterion for choosing IMFs to separate an EOG-related component from the observed signal. Numerical examples confirm the proposed approach extracts EOG component better comparing to conventional blind source separation methods.

Automatic Life-Logging: A novel approach to sense real-world activities by environmental sound cues and common sense

There are many studies that collect and store life log for personal memory. The paper explains how a system can create someones life log in an inexpensive way to share daily life events with family or friends through socialnetwork or messaging. In the modern world where people are usually busier than ever, family members are geographically distributed due to globalization of companies and humans are inundated with more information than they can process, ambient communications through mobile media or internet based communication can provide rich social connections to friends and family. People can stay connected to their loving ones ubiquitously that they care about by sharing awareness information in a passive way. For users who wish to have a persistent existence in a virtual world - to let their friends know about their current activity or to inform their caretakers - new technology is needed. Research that aims to bridge real life and the virtual worlds (e.g., Second Life, Face book etc.) to simulate virtual living or logging daily events, while challenging and promising, is currently rare. Only very recently the mapping of real-world activities to virtual worlds has been attempted by processing multiple sensors data along with inference logic for realworld activities. Detecting or inferring human activity using such simple sensor data is often inaccurate, insufficient and expensive. Hence, this paper proposes to infer human activity from environmental sound cues and common sense knowledge, which is an inexpensive alternative to other sensors (e.g., accelerometers) based approaches. Because of their ubiquity, we believe that mobile phones or hand-held devices (HHD) are ideal channels to achieve a seamless integration between the physical and virtual worlds. Therefore, the paper presents a prototype to log daily events by a mobile phone based application by inferring activities from environmental sound cues. To the best of our knowledge, this system pioneers the use of environmental sound based activity recognition in mobile computing to reflect ones real-world activity in virtual worlds.

A robust digital audio watermarking algorithm using Empirical Mode Decomposition

In this paper, we proposed a new algorithm for digital audio watermarking process using Empirical Mode Decomposition (EMD) and Hilbert Transform (HT). The host audio signal is decomposed into several Intrinsic Mode Functions (IMFs). A set of −1 and 1, which is obtained by mapping from standard normal distributed pseudo random numbers, is embedded as secret information into the IMF containing highest energy. Thus selected IMF is less sensitive with common signal processing attack. As a result this method increases robustness. The experimental results show that the proposed method has good imperceptibility and robustness under common signal processing attacks such as additive noise (in time domain and in frequency domain), low pass filtering, re-sampling, re-quantization, MP3 compression, and sound processing effects such as, delay, a natural sounding reverberation (Schroeders Reverberator), flanging effect, equalization effect.

Adaptive thresholding approach for robust voiced/unvoiced classification

This paper presents a robust voiced/unvoiced classification method by using linear model of empirical mode decomposition (EMD) controlled by Hurst exponent. EMD decomposes any signals into a finite number of band limited signals called intrinsic mode functions (IMFs). It is assumed that voiced speech signal is composed of trend due to vocal cord vibration and some noise. No trend is present in unvoiced speech signal. A linear model is developed using IMFs of the noise part of the speech signal. Then a specified confidence interval of the linear model is set as the data adaptive energy threshold. If there exists at least one IMF exceeding the threshold and its fundamental period is within the pitch range, the speech is classified as voiced and unvoiced otherwise. The experimental results show that the proposed method performs superior compared to the recently developed voiced/unvoiced classification algorithms with noticeable performance.

predCar-site: Carbonylation sites prediction in proteins using support vector machine with resolving data imbalanced issue

The carbonylation is found as an irreversible post-translational modification and considered a biomarker of oxidative stress. It plays major role not only in orchestrating various biological processes but also associated with some diseases such as Alzheimers disease, diabetes, and Parkinsons disease. However, since the experimental technologies are costly and time-consuming to detect the carbonylation sites in proteins, an accurate computational method for predicting carbonylation sites is an urgent issue which can be useful for drug development. In this study, a novel computational tool termed predCar-Site has been developed to predict protein carbonylation sites by (1) incorporating the sequence-coupled information into the general pseudo amino acid composition, (2) balancing the effect of skewed training dataset by Different Error Costs method, and (3) constructing a predictor using support vector machine as classifier. This predCar-Site predictor achieves an average AUC (area under curve) score of 0.9959, 0.9999, 1, and 0.9997 in predicting the carbonylation sites of K, P, R, and T, respectively. All of the experimental results along with AUC are found from the average of 5 complete runs of the 10-fold cross-validation and those results indicate significantly better performance than existing predictors. A user-friendly web server of predCar-Site is available at http://research.ru.ac.bd/predCar-Site/.

Window based clothing invariant gait recognition

Appearance based gait recognition becomes more difficult due to changing the gait styles by different cofactors like as cloths, carrying objects, view angles, surfaces and shoes. Out of others clothes is the most challenging issues in this area. Different part based approaches have been defined several effective and redundant body parts which can influence for individual recognition. In this paper we have study the gait by splitting it into very small window chunks and define a random window subspace method (RWSM) for clothing invariant Human gait recognition. Experiments are conducted on large-scale clothing variations OUR TEADMILL dataset B and shows superb performance than others classical gait recognition approaches.