Neethu Mohan

A VMD based approach for speech enhancement

This paper proposes a Variational Mode Decomposition (VMD) based approach for enhancement of speech signals distorted by white Gaussian noise. VMD is a data adaptive method which decomposes the signal into intrinsic mode functions (IMFs) by using the Alternating Direction Method of Multipliers (ADMM). Each IMF or mode will contain a center frequency and its harmonics. This paper tries to explore VMD as a Speech enhancement technique. In the proposed method, the noisy speech signal is decomposed into IMFs using VMD. The noisy IMFs are enhanced using two methods; VMD based wavelet shrinkage (VMD-WS) and VMD based MMSE log STSA (VMD-MMSE). The speech signal distorted with different noise levels are enhanced using the VMD based methods. The level of noise reduction and speech signal quality are measured using the objective quality measures.

Condition Monitoring in Roller Bearings using Cyclostationary Features

Proper machine condition monitoring is really crucial for any industrial and mechanical systems. The efficiency of mechanical systems greatly relies on rotating components like shaft, bearing and rotor. This paper focuses on detecting different fault in the roller bearings by casting the problem as machine learning based pattern classification problem. The different bearing fault conditions considered are, bearing-good condition, bearing with inner race fault, bearing with outer race fault and bearing with inner and outer race fault. Earlier the statistical features of the vibration signals were used for the classification task. In this paper, the cyclostationary behavior of the vibration signals is exploited for the purpose. In the feature space the vibration signals are represented by cyclostationary feature vectors extracted from it. The features thus extracted were trained and tested using pattern classification algorithms like decision tree J48, Sequential Minimum Optimization (SMO) and Regularized Least Square (RLS) based classification and provides a comparison on accuracies of each method in detecting faults.

A robust non local means maximum likelihood estimation method for Rician noise reduction in MR images

Denoising is one of the most important preprocessing task in medical image analysis. It has a great role in the clinical diagnosis and computerized analysis. When SNR is low, medical images follows a Rician noise distribution which is signal dependent. In the literature, only few works focus on the edge preserving quality of MR images. Our aim is to estimate the noise free signal from MR magnitude images by focusing on preserving edges and tissue boundaries. The proposed method is an improvisation over non local means maximum likelihood approach for Rician noise reduction in MR images. Our method focus on a robust estimator function (Geman-McClure function) for weight calculation, and is compared with the existing methods in terms of PSNR ratio, visual quality comparison and by SSIM values. The proposed method outperforms the state-of-the art methods in preserving fine structural details and edge boundaries.

An \({\ell _1}\)-Norm Based Optimization Approach for Power Line Interference Removal in ECG Signals

Accurate analysis and proper interpretation of electrophysical recordings like ECG is a real necessity in medical diagnosis. Presence of artifacts and other noises can corrupt the ECG signals and can lead to an improper disease diagnosis. Power line interferences (PLI) occurring at 50/60 Hz is a major source of noises which could corrupt the ECG signals. This motivates the removal of PLI from ECG signals and is a foremost preprocessing task in ECG signal analysis. In this paper, we deal an \({\ell _1}\) norm based optimization approach for PLI removal in ECG signals. The sparsity inducing property of \({\ell _1}\) norm is used for efficient removal of power noises. The effectiveness of this approach is evaluated on ECG signals corrupted with power line interferences and random noises.

Performance evaluation of edge feature extracted using sparse banded matrix filter applied for face recognition

This paper deals with the performance evaluation of sparse banded matrix filter applied for Face recognition. Edges extracted using the sparse banded matrix filter (ABFilter) is used as a feature descriptor for face recognition. The classification is done using Random Kitchen Sink which is accessed through GURLS library and also classified using Support Vector Machines (SVM). The experimental evaluation of sparse banded matrix filter is done on a standard face database (Yale). Edge detection is the process of locating the sharp discontinuity in an image. It is a basic tool which is used in many image processing applications such as face recognition. In this paper, we have compared the performance of sparse banded matrix filter with existing edge detecting filters such as Sobel, Prewitt, Canny and Robert. Though many filters exist for edge detection, sparse banded matrix filter is known for the edge detection with minimal discontinuity. The experimental evaluation shows that the edge feature descriptors of Yale face database obtained using sparse banded matrix filter provides 88 % accuracy using GURLS and 81% using SVM.

Denoising of phonocardiogram signals using variational mode decomposition

Recent advances in signal processing and the revolution by the mobile technologies have spurred several innovations in all the areas and albeit more so in home based tele-medicine. We used variational mode decomposition (VMD) based denoising on large-scale phonocardiogram (PCG) data sets and achieved better accuracy. We have also implemented a reliable, external hardware and mobile based phonocardiography system that uses VMD signal processing technique to denoise the PCG signal that visually displays the waveform and inform the end-user and send the data to cloud based analytics system.

Least square based image deblurring

An image can be basically defined as an object that represents visual observation, which can be created and stored in the electronic form, produced from an optical device. When we take a photograph, there can be many problems associated with that particular image. Among them, one of the main issue is the blur of the image. Blur can be defined as something which will become vague or less distinct. A blurred image looks sharper or more detailed, if we are able to perceive all the objects and their shapes correctly in it. The main cause for blur is the out of focus issue of the camera/sensor. An image which is in out of focus will appear in a blurred state. Even if, at the present time, with an auto focus facility, sometimes we will not get the image in the correct focus. Most probably, a part of the image will be crisp and clear, however rest will be ill-defined. Image deblurring is a common and important process in fields like digital photography, medical imaging and astronomy. Hence, removing or dropping the total amount of blur is the most important task before being applying to the image analysis techniques. In this paper, a colour image deblurring algorithm based on the concept of least squares is proposed. The 1D least square based deconvolution technique is extended to colour image deblurring. The proposed approach is experimented on standard test images and the results are compared with classical total variation image deblurring algorithm. The effectiveness of the proposed approach is evaluated in terms of standard quality metrics such as PSNR and SSIM.

A Novel Cyclic Convolution Based Regularization Method for Power-Line Interference Removal in ECG Signal.

Applying signal processing to bio-signal record such as electrocardiogram or ECG signals provide vital insights to the details in diagnosis. The diagnosis will be exact when the extracted information about the ECG is accurate. However, these records usually gets corrupted/contaminated with several artifacts and power-line interferences (PLI) thereby affects the quality of diagnosis. Power-line interferences occurs in the range close to 50 Hz/60 Hz. The challenge is to remove the interferences without altering the original characteristics of ECG signal. Since the ECG signals frequency range is close to PLI, several articles discuss PLI removal methods which are mathematically complex and computationally intense. The present paper proposes a novel PLI removal method that uses a simple optimization method involving a circular convolution based \({\ell _2}\)-norm regularization. The solution is obtained in closed form and hence computationally simple and fast. The effectiveness of the proposed method is evaluated using output signal-to-noise-ratio (SNR) measure, and is found to be state-of-the-art.

Overlapping Group Sparsity Induced Condition Monitoring in Rotating Machineries.

Maintenance of rotating parts in machines is not easy. Prediction of faults in advance reduces the frequency of breakdown and improves the life time of machines. This paper proposes a machine condition monitoring system, which formulates the fault diagnosis problem as a machine learning based pattern classification problem. The vibration signals acquired from rotating machines are initially processed by a group-sparse denoising algorithm namely Overlapping Group Shrinkage (OGS). In OGS, the group sparse signal denoising problem is casted as a convex optimization problem with a group sparsity promoting penalty function. The denoised signals are then processed by Variational Mode Decomposition (VMD), which decomposes the signal into specific frequency modes. For representing the signal in the feature space, energy of each mode is extracted and is classified by LS-SVM classifier. The performance of the proposed condition monitoring system is evaluated in terms of classification accuracies and is compared with statistical features.

Accurate frequency estimation method based on basis approach and empirical wavelet transform

Due to proliferating harmonic pollution in the power system, analysis and monitoring of harmonic variation in real-time have become important. In this paper, a novel approach for estimation of fundamental frequency in power system is discussed. In this method, the fundamental frequency component of the signal is extracted using Empirical Wavelet Transform. The extracted component is then projected onto fourier basis, where the frequency is estimated to a resolution of 0.001 Hz. The proposed approach gives an accurate frequency estimate compared with some existing methods.