WSEAS Transactions on
Signal Processing

2016 | 2015 | 2014 | 2013 | 2012 | 2011 | 2010 | 2009 | 2008 | Pre-2008

Print ISSN: 1790-5052
E-ISSN: 2224-3488

Volume 8, 2012

Issue 1, Volume 8, January 2012

Title of the Paper: Bound the Learning Rates with Generalized Gradients

Authors: Sheng Baohuai, Xiang Daohong

Abstract: This paper considers the error bounds for the coef cient regularized regression schemes associated with Lipschitz loss. Our main goal is to study the convergence rates for this algorithm with non-smooth analysis. We give an explicit expression of the solution with generalized gradients of the loss which induces a capacity independent bound for the sample error. A kind of approximation error is provided with possibility theory.

Keywords: Regularization regression, non-smooth analysis, Lipschitz loss, machine learning, learning rates, generalized gradient

Title of the Paper: The Voice Segment Type Determination using the Autocorrelation Compared to Cepstral Method

Authors: Old?ich Horák

Abstract: The extraction of the characteristic features of the speech is the important task in the speaker recognition process. One of the basic features is fundamental frequency of speaker’s voice, which can be extracted from the voiced segment of the speech signal. This document describes one of the methods providing possibility to distinguish the voiced and surd segments of the voice signal using the autocorrelation, and compare the results to cepstral method.

Keywords: autocorrelation, cepstrum, features extraction, fundamental frequency, signal processing, speaker recognition, voice signal

Title of the Paper: Facial Expression Recognition Based on Local Binary Patterns and Local Fisher Discriminant Analysis

Authors: Shiqing Zhang, Xiaoming Zhao, Bicheng Lei

Abstract: Automatic facial expression recognition is an interesting and challenging subject in signal processing, pattern recognition, artificial intelligence, etc. In this paper, a new method of facial expression recognition based on local binary patterns (LBP) and local Fisher discriminant analysis (LFDA) is presented. The LBP features are firstly extracted from the original facial expression images. Then LFDA is used to produce the low dimensional discriminative embedded data representations from the extracted high dimensional LBP features with striking performance improvement on facial expression recognition tasks. Finally, support vector machines (SVM) classifier is used for facial expression classification. The experimental results on the popular JAFFE facial expression database demonstrate that the presented facial expression recognition method based on LBP and LFDA obtains the best recognition accuracy of 90.7% with 11 reduced features, outperforming the other used methods such as principal component analysis (PCA), linear discriminant analysis (LDA), locality preserving projection (LPP).

Keywords: Facial expression recognition, local binary patterns, local Fisher discriminant analysis, support vector machines, principal component analysis, linear discriminant analysis, locality preserving projection

Title of the Paper: Combined Fuzzy Logic and Unsymmetric Trimmed Median Filter Approach for the Removal of High Density Impulse Noise

Authors: T. Veerakumar, S. Esakkirajan, Ila Vennila

Abstract: In this paper, a combined fuzzy logic and unsymmetric trimmed median filter approach is proposed to remove the high density salt and pepper noise in gray scale and colour images. This algorithm is a combination of decision based unsymmetrical trimmed median filter and fuzzy thresholding technique to preserve edges and fine details in an image. The decision based unsymmetric trimmed median filter fails if all the elements in the selected window are 0’s or 255’s. One of the possible solutions is to replace the processing pixel by the mean value of the elements in the window. This will lead to blurring of the edges and fine details in the image. To preserve the edges and fine details, the combined fuzzy logic and unsymmetric trimmed median filter approach is proposed in this paper. The better performance of the proposed algorithm is demonstrated on the basis of PSNR and IEF values.

Keywords: Fuzzy logic, Fuzzy threshold, Salt and Pepper noise, Decision based Unsymmetric Trimmed Median Filter, Membership function, Noise reduction

Issue 2, Volume 8, April 2012

Title of the Paper: A Lossless Image Compression Algorithm Using Predictive Coding Based on Quantized Colors

Authors: Fuangfar Pensiri, Surapong Auwatanamongkol

Abstract: Predictive coding has proven to be effective for lossless image compression. Predictive coding estimates a pixel color value based on the pixel color values of its neighboring pixels. To enhance the accuracy of the estimation, we propose a new and simple predictive coding that estimates the pixel color value based on the quantized pixel colors of three neighboring pixels. The prediction scheme can help minimize the upper bound of the residual errors from the prediction. The experiments cover a set of true color 24-bit images, whose pixel colors are quantized into 2, 4, 8 and 16 colors. The results show that the proposed algorithm outperforms some well known lossless image compression algorithms such as JPEG-LS and PNG by factors of 2-3 in terms of bits per pixel. The results also show that the proposed coding gives the best compression rates when colors are quantized into two colors.

Keywords: Image compression, Lossless compression, Lossless image compression, Compression, Predictive coding, Quantized colors

Title of the Paper: A New Feature Reduction Method and Its Application in the Reciprocating Engine Fault Diagnosis

Authors: Ma Jin, Jiang Zhinong

Abstract: On the basis of complicated fault feature of the reciprocating engine, a new feature reduction method based on the principle of the knowledge granularity to estimate the significance of symptomatic parameters is presented in this paper. The current problem that in the process of reducing and compressing the symptomatic parameters of fault diagnosis, the smallest symptom sets obtained is not always the smallest and optimal one, has been solved by the new method. By calculating on two instance of reciprocating engine knowledge set, the feature reduction method is effective.

Keywords: symptomatic parameter, reciprocating engine, granularity entropy, fault diagnosis, fault feature, knowledge granularity

Title of the Paper: An Adaptive Matrix Embedding Technique for Binary Hiding With an Efficient LIAE Algorithm

Authors: Jyun-Jie Wang, Houshou Chen, Chi-Yuan Lin

Abstract: Researchers have developed a great number of embedding techniques in steganography. Matrix embedding, otherwise called the binning scheme, is one such technique that has been proven to be an efficient algorithm. Unlike conventional matrix embedding, which requires a maximum likelihood decoding algorithm to find the coset leader, this study proposes an adaptive algorithm called the linear independent approximation embedding (LIAE) algorithm. There are numerous concerns with the cover location selection, such as less significant cover to be modified, alterable part of the cover and forced the cover to be modified, when embedding a secret message into the cover. The LIAE algorithm has the ability to perform data embedding at an arbitrarily specified cover location. Therefore, the embedded message can be identified at the receiver without incurring any damage to the associated cover location. The simulation results show that the LIAE embedding algorithm has superior efficiency and adaptability compared with other suboptimal embedding algorithms. Moreover, the experimental results also demonstrate the trade-off between embedding efficiency and computational complexity.

Keywords: Steganography, matrix embedding, ML decoding, coset leader, embedding efficiency, linear block code

Issue 3, Volume 8, July 2012

Title of the Paper: A New Approach to Diagnose Induction Motor Defects based on the Combination of the TSA Method and MCSA Technique

Authors: Nabil Ngote, Saïd Guedira, Mohamed Cherkaoui

Abstract: In this article, we are trying to exploit the cyclostationary characteristics of electrical signals in order to detect the rotor faults of an asynchronous machine. These defects are the most complex in terms of detection since they interact with the 50 Hz carrier with a weak band occupied in frequency. The test bench used includes an industrial three-phase wound rotor asynchronous motor of 400V, 6.2A, 50Hz, 3kW, 1385rpm characteristics (Fig. 15). The rotor fault has been carried out by adding an extra 40m? resistance on one of the rotor phases (i.e. 10% of the rotor resistance value per phase, Rr=0,4?). From the stator voltage and current acquisition, and by application of the Time Synchronous Averaging (TSA) method to the stator current, we condition the electrical signal in order to obtain a sensitive indicator allowing to easily distinguish the healthy cases from defective ones; this indicator will allow the motor monitoring. In a second step, we will apply the Motor Current Signature Analysis (MCSA) technique to the stator current, in order to identify the type of the detected fault. This will allow to go further and diagnose the motor defect.

Keywords: Cyclostationarity; Time Synchronous Averaging (TSA); Monitoring; Rotor fault; Spectral analysis; Motor Current Signature Analysis (MCSA)

Title of the Paper: An Optimal EEG-based Emotion Recognition Algorithm Using Gabor Features

Authors: Saadat Nasehi, Hossein Pourghassem

Abstract: Feature extraction and accurate classification of the emotion-related EEG-characteristics have a key role in success of emotion recognition systems. In this paper, an optimal EEG-based emotion recognition algorithm based on spectral features and neural network classifiers is proposed. In this algorithm, spectral, spatial and temporal features are selected from the emotion-related EEG signals by applying Gabor functions and wavelet transform. Then neural network classifiers such as improved particle swarm optimization (IPSO) and probabilistic neural network (PNN) are developed to determine an optimal nonlinear decision boundary between the extracted features from the six basic emotions (happiness, surprise, anger, fear, disgust and sadness). The best result is obtained when Gabor-based features and PNN classifier are used. In this condition, our algorithm can achieve average accuracy of 64.78% that can be used in brain-computer interfaces systems.

Keywords: Electroencephalogram, emotion recognition, wavelet transform, Gabor functions, improved particle swarm optimization (IPSO), probabilistic neural network (PNN)

Title of the Paper: Signal Detection for OFDM and DS-CDMA with Gradient and Blind Source Separation Principles

Authors: M. G. S. Sriyananda, J. Joutsensalo, T. Hamalainen

Abstract: Signal recovery mechanisms for both Orthogonal Frequency Division Multiplexing (OFDM) and Direct Sequence - Code Division Multiple Access (DS-CDMA) with the assistance of principles of Blind Source Separation (BSS) and Gradient Algorithms (GAs) are presented. Elimination or reduction of undesirable influences encountered with in the wireless interface is targeted using a set of filter coefficients. Four energy functions are used in deriving them. Time correlation properties of the channel are used as advantages in introducing the energy functions and they are tried to be justified followed by a performance evaluation. All the schemes are tested with synchronous downlink system models. Simulations are carried out under slow fading channel conditions with a receiver containing Equal Gain Combining (EGC) and BSS algorithms. It could be noted that, better performance for this predominant air interface communication techniques can be achieved with this combined schemes. It is important grasp the fact that, these schemes can be promoted as low complexity simple matrices based processing mechanisms.

Keywords: Blind Source Separation, Gradient Algorithms, OFDM, DS-CDMA, Slow Fading

Title of the Paper: Novel Particle Swarm Optimization for Low Pass FIR Filter Design

Authors: Sangeeta Mondal, S. P. Ghoshal, Rajib Kar, Durbadal Mandal

Abstract: This paper presents an optimal design of linear phase digital low pass finite impulse response (FIR) filter using Novel Particle Swarm Optimization (NPSO). NPSO is an improved particle swarm optimization (PSO) that proposes a new definition for the velocity vector and swarm updating and hence the solution quality is improved. The inertia weight has been modified in the PSO to enhance its search capability that leads to a higher probability of obtaining the global optimal solution. The key feature of the proposed modified inertia weight mechanism is to monitor the weights of particles, which linearly decrease in general applications. In the design process, the filter length, pass band and stop band frequencies, feasible pass band and stop band ripple sizes are specified. FIR filter design is a multi-modal optimization problem. Evolutionary algorithms like real code genetic algorithm (RGA), particle swarm optimization (PSO), and the novel particle swarm optimization (NPSO) have been used in this work for the design of linear phase FIR low pass (LP) filter. A comparison of simulation results reveals the optimization efficacy of the algorithm over the prevailing optimization techniques for the solution of the multimodal, non-differentiable, highly non-linear, and constrained FIR filter design problems.

Keywords: FIR Filter, RGA, PSO, NPSO, Parks and McClellan (PM) Algorithm, Evolutionary Optimization, Low Pass Filter

Title of the Paper: Heuristic Search Method for Digital IIR Filter Design

Authors: Ranjit Kaur, Manjeet Singh Patterh, J. S. Dhillon, Damanpreet Singh

Abstract: The paper develops innovative methodology for robust and stable design of digital infinite impulse response (IIR) filters using a heuristic search method. The proposed heuristic search method enhances the capability to explore and exploit the search space locally as well globally to obtain optimal filter design parameters. A multicriterion optimization is employed as design criterion to obtain optimal stable IIR filter that satisfies different performance requirements like minimizing Lp-norm approximation error and ripple magnitude. Multicriterion optimization problem has been solved applying weighted sum method and p-norm method. Best weight pattern is searched using evolutionary search method that minimizes the performance criteria simultaneously. The proposed heuristic search method is effectively applied to solve the multicriterion, multiparameter optimization problems of low-pass, high-pass, band-pass, and band-stop digital filters design. The computational experiments show that the proposed heuristic search method is superior or atleast comparable to other algorithms and can be efficiently used for higher filter design.

Keywords: Digital infinite impulse response filters, Heuristic search algorithm, Multicriterion optimization, Magnitude Response, Stability, Lp-norm error

Title of the Paper: Efficient Algorithm for Enhancement of Images Corrupted by Salt & Pepper Noise

Authors: Shyam Lal, Mahesh Chandra

Abstract: In this paper an efficient algorithm is proposed for removal of salt & pepper noise from digital images. The proposed algorithm consists of two stages: in first stage the noisy image is processed by adaptive median filter and in the second stage the output of first stage is further processed by modified mean filter. The first stage classifies noisy pixels by comparing each pixel in the image to its surrounding neighbourhood pixels. If the pixel is different from a majority of its neighbours, then it is considered as one of the noisy pixels. Noisy pixels are replaced by the median value of the neighbourhood pixels. Second stage works in two phases: in the first phase the noisy pixels are detected and in second phase each noisy pixel are replaced by the mean value of noise free pixel of 2×2 window. Simulation and experimental results show that the proposed algorithm consistently works well in suppressing the salt and pepper noise density (up to 95%). The proposed Adaptive Median based modified Mean filter (AMMF) outperforms a number of other existing filters such as standard median filter (SMF), centre weighted median filter (CWMF), progressive switching median filter (PSMF), open-close sequence filter (OCSF), decision based algorithm (DBA), modified decision based unsymmetric trimmed median filter (MDBUTMF) for noise removal in terms of high PSNR, low MSE and reduced streaking effect. The proposed algorithm is suitable for low level noise density as well as high level noise density. The proposed algorithm demonstrates better performance as compared to other existing techniques on different gray scale images.

Keywords: Median filter, Centre weighted median filter, Open-close sequence filter, Decision based algorithm, Modified decision based unsymmetric trimmed median filter

Issue 4, Volume 8, October 2012

Title of the Paper: Fusion of Infrared and Visual Images Using Bacterial Foraging Strategy

Authors: Rutuparna Panda, Manoj Kumar Naik

Abstract: This paper presents new methods for fusion of the visual and thermal images for pattern recognition. Researchers have suggested different fusion schemes to find out pattern vectors for object detection and recognition. The different fusion schemes are — data fusion, decision fusion etc. These schemes have been proposed in different way to improve the performance. Hence, here we propose three new methods for fusing the visual and infrared (IR) images. The proposed new methods are – fusion using information content from Gray Level Co-occurrence Matrix (GLCM), fusion using wavelet energy signature and fusion by maximizing wavelet energy signature using E. coli bacteria foraging strategy (EBFS). In the third method, we consider information fusion as an optimization problem and then solve it using EBFS as a search algorithm. Finally, we compare the results using the contrast signature from GLCM and observed that the later scheme using EBFS shows better results than other two methods.

Keywords: Pattern recognition, Evolutionary computation, Bacteria foraging, Wavelet theory

Title of the Paper: About Fourier Transform

Authors: Luiza Grigorescu, Gheorghe Oproescu, Ioana Diaconescu

Abstract: This paper analyses Fourier transform used for spectral analysis of periodical signals and emphasizes some of its properties. It is demonstrated that the spectrum is strongly depended of signal duration that is very important for very short signals which have a very rich spectrum, even for totally harmonic signals. Surprisingly is taken the conclusion that spectral function of harmonic signals with infinite duration is identically with Dirac function and more of this no matter of duration, it respects Heisenberg fourth uncertainty equation. In comparison with Fourier series, the spectrum which is emphasized by Fourier transform doesn’t have maximum amplitudes for signals frequencies but only if the signal lasting a lot of time, in the other situations these maximum values are strongly de-phased while the signal time decreasing. That is why one can consider that Fourier series is useful especially for interpolation of non-harmonic periodical functions using harmonic functions and less for spectral analysis.

Keywords: Signals, Fourier transform, continuous spectrum properties, Quantum Physics, Fourier series, discrete spectrum

Title of the Paper: Boundary Effects Reduction in Wavelet Transform for Time-frequency Analysis

Authors: Hang Su, Quan Liu, Jingsong Li

Abstract: Boundary effects are very common in the processing of finite-length signals. In this paper, we consider the problem of handling the boundary effects that can occur due to improper extension methods. Contrary to traditional methods including zero padding, periodic extension and symmetric extension, we propose an extension algorithm based on Fourier series with properties that make it more suitable for boundary effects reduction in the application of time-frequency signal analysis. This extension algorithm could preserve the time-varying characteristics of the signals and be effective to reduce artificial singularities appearing at the boundary. Procedures for realization of the proposed algorithm and relative issues are presented. Accurate expressions for the extent of boundary effects region are derived and show that the extent of boundary effects region is not equivalent to the width of wavelet under current mean square definition. Then, an interpolation approach is used in the boundary effects region to further alleviate the distortions. Several experimental tests conducted on synthetic signals exhibiting linear and nonlinear laws are shown that the proposed algorithms are confirmed to be efficient to alleviate the boundary effects in comparison to the existing extension methods.

Keywords: Finite-length Signals, Convolution, Wavelet Transform, Boundary Effects, Fourier Series Extension, Interpolation

Title of the Paper: Unearthing Clues to Reduce the Devastating Effects of Earthquakes: The Hilbert-Huang Transform

Authors: Silvia Garcia, Miguel Romo

Abstract: In this paper the use of the energy-time-frequency representation, the Hilbert-Huang Transform HHT, for the decomposition and characterization of seismic ground response signals, is discussed. The HHT, integrated by the Empirical Mode Decomposition (EMD) and the Hilbert Transformation (HT), enables engineers to analyze non-stationary oscillation systems and to obtain more detailed intensity descriptions on time-varying frequency diagrams. The advantage of the HHT over other representations is its sharp intensity-localization properties in the time-frequency plane. This paper first provides the fundamentals of the HHT method, and then uses them to analyze recordings of Mexico City soft-soil deposits, indicating that the HHT method is able to extract some motion characteristics useful in studies of seismology and geotechnical engineering, which might not be exposed effectively and efficiently by other conventional data processing techniques.

Keywords: Hilbert-Huang Transform, Seismic recordings, Site effects, Time Series Analysis