Search

Problem statement: This study proposed an approach for classification of placental tissues development using ultrasound images. Approach: This approach was based to the selection of tissues, feature extraction by discrete wavelet transform and classification by neural network and especially the Multi Layer Perceptron (MLP). Results: The proposed approach was tested for ultrasound placental images; resulting in 95% success rate. Conclusion/Recommendations: The method showed a good recognition for placental tissues and will be useful for detection of the placental anomalies those concerning the premature birth and the intrauterine growth retardation. Key words: Placenta, wavelet transform, neural network, MLP, INTRODUCTION An ultrasound diagnostic system has become an important and popular diagnostic tool since it has a wide range of applications. Specifically, due to its noninvasive and non-destructive nature, the [...]

We investigate image and video denoising using adaptive dual-tree discrete wavelet packets (ADDWP), which is extended from the dual-tree discrete wavelet transform (DDWT). With ADDWP, DDWT subbands are further decomposed into wavelet packets with anisotropic decomposition, so that the resulting wavelets have elongated support regions and more orientations than DDWT wavelets. To determine the decomposition structure, we develop a greedy basis selection algorithm for ADDWP, which has significantly lower computational complexity than a previously developed optimal basis selection algorithm, with only slight performance loss. For denoising the ADDWP coefficients, a statistical model is used to exploit the dependency between the real and imaginary parts of the coefficients. The proposed denoising scheme gives better performance than several state-of-the-art DDWT-based schemes for images with rich directional features. Moreover, our scheme shows promising results without using motion estimation in video denoising. The visual quality of images and videos denoised by the proposed scheme is also superior. Index Terms--Anisotropic decomposition, complex wavelet packets, directional transform, image denoising, video denoising.

A novel hybrid protection algorithm, based on traveling wave protection principle and boundary protection principle for a monopolar HVDC line is proposed. Stationary wavelet transform (SWT) is adopted in the traveling wave protection to process the dc signal and then wavelet modulus maxima are used to further represent the useful traveling wave signal. The boundary protection principle based on SWT is used jointly with traveling wave protection to distinguish internal faults from external faults. The effect of border distortion, noise, high ground fault resistance, close-up faults, transients caused by lightning strokes and different dc line terminations are considered in the paper. Index Terms--Boundary protection, HVDC line protection, lightning strokes, relay, stationary wavelet transform (SWT), traveling waves.

Accurate fault classification and swift faulted-phase selection are the bases of protection for transmission systems. This paper presents the algorithm of fault classification and faulted-phase selection based on the initial current traveling wave. The characteristics of various faults are investigated on the basis of the Karenbauer transform. The criteria of fault classification and faulted-phase selection are put forward. In order to extract the traveling wave from postfault signals and to construct the algorithm, wavelet transform technology is adopted. The simulations based on the electromagnetic transients program have been done to verify the performance of the proposed algorithm. Index Terms--Dyadic wavelet transform, fault classification, faulted-phase selection, initial current traveling wave, modulus maxima.

The present work describes a Bayesian maximum a posteriori (MAP) method using a statistical multiscale wavelet prior model. Rather than using the orthogonal discrete wavelet transform (DWT), this prior is built on the translation invariant wavelet transform (TIWT). The statistical modeling of wavelet coefficients relies on the generalized Gaussian distribution. Image reconstruction is performed in spatial domain with a fast block sequential iteration algorithm. We study theoretically the TIWT MAP method by analyzing the Hessian of the prior function to provide some insights on noise and resolution properties of image reconstruction. We adapt the key concept of local shift invariance and explore how the TIWT MAP algorithm behaves with different scales. It is also shown that larger support wavelet filters do not offer better performance in contrast recovery studies. These theoretical developments are confirmed through simulation studies. The results show that the proposed method is more attractive than other MAP methods using either the conventional Gibbs prior or the DWT-based wavelet prior. Index Terms--Emission tomography, expectation-maximization, image reconstruction, maximum a posteriori, translation invariant, wavelet transform.

In this brief, by combining an efficient wavelet representation with a coupled map lattice model, a new family of adaptive wavelet neural networks, called lattice dynamical wavelet neural networks (LDWNNs), is introduced for spatio--temporal system identification. A new orthogonal projection pursuit (OPP) method, coupled with a particle swarm optimization (PSO) algorithm, is proposed for augmenting the proposed network. A novel two-stage hybrid training scheme is developed for constructing a parsimonious network model. In the first stage, by applying the OPP algorithm, significant wavelet neurons are adaptively and successively recruited into the network, where adjustable parameters of the associated wavelet neurons are optimized using a particle swarm optimizer. The resultant network model, obtained in the first stage, however, may be redundant. In the second stage, an orthogonal least squares algorithm is then applied to refine and improve the initially trained network by removing redundant wavelet neurons from the network. An example for a real spatio--temporal system identification problem is presented to demonstrate the performance of the proposed new modeling framework. Index Terms--Coupled map lattice, neural networks, particle swarm optimization (PSO), spatio--temporal systems, wavelets.

This paper presents the practical implementation of a novel fault diagnostic and protection scheme for the interior permanent-magnet (IPM) synchronous motors using wavelet packet transform (WPT) and artificial neural network. In the proposed technique, the line currents of different faulted and normal conditions of the IPM motor are preprocessed by the WPT. The second level WPT coefficients of line currents are used as inputs of a three-layer feedforward neural network. The proposed protection technique is successfully simulated and experimentally tested on the line-fed and inverter-fed IPM motors. The Texas Instrument 32-bit floating-point digital signal processor TMS320C31 is used for the real-time implementation of the proposed protection algorithm. The offline and online test results of both line-fed and inverter-fed IPM motors are given. These test results showed satisfactory performances of the proposed diagnostic and protection technique in terms of speed, accuracy, and reliability. Index Terms--Digital signal processors, fault diagnosis, interior permanent-magnet (IPM) motor, neural networks, wavelet packet transform (WPT).

A novel data structure is proposed for magnitude-ordering 3-D wavelet transform coefficients. This data structure consists of temporal 1-D orientation trees followed by spatial 2-D orientation trees. Based on this data structure, a coding algorithm is developed for the embedded coding of 3-D wavelet transform coefficients. Experiment results confirm that the proposed coding algorithm outperforms a conventional algorithm that uses asymmetric 3-D orientation trees. Index Terms--Embedded coding, set partitioning in hierarchical trees (SPIHT), video coding, wavelet transform.

Majorization-minimization (MM) optimization problems related to wavelet-based image restoration are examined. The iteratively reweighted shrinkage (IRS-2) algorithm is best for problems involving severe blurs and the iterative shrinkage thresholding (IST) algorithm in problems involving mild blur and medium to large noise.

This paper proposes a novel wavelet-transform-based directional algorithm for busbar protection. The algorithm decomposes the current and voltage signals into their first-level details, which consist of frequencies in 500- to 1000-Hz bandwidth, for generating directional signals. A high level of computational efficiency is achieved compared to the other wavelet-transform-based algorithms proposed, since only the high-frequency details at the first level are employed in this algorithm. The validity of this method was exhaustively tested by simulating various types of faults on a substation modeled in the Alternative Transients Program/Electromagnetic Transients Program. The algorithm correctly discriminated between bus faults, various types of external faults, and transformer energization even in the presence of current-transformer saturation. This paper also provides the design details of the algorithm using field-programmable gate array technology. Index Terms--Busbar protection, detail coefficients, digital directional relay, field-programmable gate array (FPGA) implementation, wavelet transforms.

This paper outlines the implementation details and enhancements of a previously described new concept in blind seismic deconvolution that is referred to as principle phase decomposition (PPD). A requirement of the PPD algorithm is for the investigator to determine the seismogram's dominant frequency (DF) and corresponding principle phase components (PPCs). Once these parameters are estimated, a hybrid Rao-Blackwellized particle filter and a hidden Markov model (HMM) filter are utilized to separate the potentially time-variant overlapping source wavelets. A variation of the PPD algorithm that is referred to as the PPD wavelet extraction (PPD-WE) technique addresses the requirement of estimating the seismogram's DF and PPCs. This paper describes in detail the PPD-WE algorithm where the overlapping source wavelets are sequentially and chronologically extracted from the seismogram under analysis. A HMM filter is described which facilitates in the simultaneous estimation of the DF and the corresponding phase of the source wavelet to be extracted within the PPD-WE algorithm. In addition, the utilization of the PPD-WE algorithm within standard frequency-domain deconvolution techniques is outlined. Index Terms--Blind deconvolution, hidden Markov models (HMMs), jump processes, Rao-Blackwellized particle filter (RBPF).

Wavelets have emerged as powerful tools for signal coding especially bio-signal processing. Wavelet transform is used to represent the signal to some other time--frequency representation better suited for detecting and removing redundancies. A novel algorithm for wavelet based ECG signal coding is proposed in this paper. Experimental results show that this algorithm outperforms than other coders such as Djohn, EZW, SPIHT, etc., exits in the literature in terms of simplicity and coding efficiency by successive partition the wavelet coefficients in the space frequency domain and send them using adaptive decimal to binary conversion. Proposed algorithm is significantly more efficient in compression, simple in implementation and in computation than the previously proposed coders. This algorithm is tested for 26 different records from MIT-BIH arrhythmia database and obtained an average percent root mean square difference as around 0.01-4.8% for an average compression ratio of 2:1 to 35:1. A compression ratio of 8.5108:1 is achieved for MIT--BIH arrhythmia database record 117 with a percent mean square difference as 1.29%. Keywords: ECG signal compression; Wavelets; Mean; Successive partitions; Adaptive decimal to binary conversion

A noninvasive and unconstrained real-time method to detect the respiration rhythm and pulse rate during sleep is presented. By employing the a trous algorithm of the wavelet transformation (WT), the respiration rhythm and pulse rate can be monitored in real-time from a pressure signal acquired with a pressure sensor placed under a pillow. The waveform for respiration rhythm detection is derived from the 26 scale approximation, while that for pulse rate detection is synthesized by combining the 24 and 25 scale details. To minimize the latency in data processing and realize the highest real-time performance, the respiration rhythm and pulse rate are estimated by using waveforms directly derived from the WT approximation and detail components without the reconstruction procedure. This method is evaluated with data collected from 13 healthy subjects. By comparing with detections from finger photoelectric plethysmograms used for pulse rate detection, the sensitivity and positive predictivity were 99.17% and 98.53%, respectively. Similarly, for respiration rhythm, compared with detections from nasal thermistor signals, results were 95.63% and 95.42%, respectively. This study suggests that the proposed method is promising to be used in a respiration rhythm and pulse rate monitor for real-time monitoring of sleep-related diseases during sleep. Index Terms--a trous algorithm, pulse rate, real-time monitor, respiration rhythm, sleep monitor, wavelet transformation.

In this paper, a new image compression algorithm is proposed based on the efficient construction of wavelet coefficient lower trees. The main contribution of the proposed lower-tree wavelet (LTW) encoder is the utilization of coefficient trees, not only as an efficient method of grouping coefficients, but also as a fast way of coding them. Thus, it presents state-of-the-art compression performance, whereas its complexity is lower than the one presented in other wavelet coders, like SPIHT and JPEG 2000. Fast execution is achieved by means of a simple two-pass coding and one-pass decoding algorithm. Moreover, its computation does not require additional lists or complex data structures, so there is no memory overhead. A formal description of the algorithm is provided, while reference software is also given. Numerical results show that our codec works faster than SPIHT and JPEG 2000 (up to three times faster than SPIHT and fifteen times faster than JPEG 2000), with similar coding efficiency. Index Terms--Image compression, low complexity, tree-based coding, wavelets.

This paper proposes the use of wavelets for the identification of an unknown sparse system whose impulse response (IR) is rich is spectral content. The superior time localization property of wavelets allows for the identification and subsequent adaptation of only the nonzero IR regions, resulting in lower complexity and faster convergence speed. An added advantage of using wavelets is their ability to partially decorrelate the input, thereby further increasing convergence speed. Good time localization of nonzero IR regions requires high temporal resolution while good decorrelation of the input requires high spectral resolution. To this end we also propose the use of biorthogonal wavelets which fulfill both of these two requirements to provide additional gain in performance. The paper begins with the development of the wavelet-basis (WB) algorithm for sparse system identification. The WB algorithm uses the wavelet decomposition at a single scale to identify the nonzero IR regions and subsequently determines the wavelet coefficients of the unknown sparse system at other scale levels that require adaptation as well. A special implementation of the WB algorithm, the successive-selection wavelet-basis (SSWB), is then introduced to further improve performance when certain a priori knowledge of the sparse IR is available. The superior performance of the proposed methods is corroborated through simulations. Index Terms--Adaptive algorithm, sparse impulse response (IR), system identification, wavelets.

This paper presents a dysphonic voice classification system using the wavelet packet transform and the best basis algorithm (BBA) as dimensionality reductor and 06 artificial neural networks (ANN) acting as specialist systems. Each ANN was a 03-layer multilayer perceptron with 64 input nodes, 01 output node and in the intermediary layer the number of neurons depends on the related training pathology group. The dysphonic voice database was separated in five pathology groups and one healthy control group. Each ANN was trained and associated with one of the 06 groups, and fed by the best base tree (BBT) nodes' entropy values, using the multiple cross validation (MCV) method and the leave-one-out (LOO) variation technique and success rates obtained were 87.5%, 95.31%, 87.5%, 100%, 96.87% and 89.06% for the groups 01 to 06, respectively. Index Terms--Acoustical analysis of voices, artificial neural network, dysphonic voice classification, wavelet packet transform.

Several studies have demonstrated the effectiveness of the Haar wavelet decomposition as a tool for reducing large amounts of data down to compact wavelet synopses that can be used to obtain fast, accurate approximate answers to user queries. Although originally designed for minimizing the overall mean-squared (i.e., L2-norm) error in the data approximation, recently proposed methods also enable the use of Haar wavelets in minimizing other error metrics, such as the relative error in data value reconstruction, which is arguably the most important for approximate query answers. Relatively little attention, however, has been paid to the problem of using wavelet synopses as an approximate query answering tool over complex tabular datasets containing multiple measures, such as those typically found in real-life OLAP applications. Existing decomposition approaches will either operate on each measure individually, or treat all measures as a vector of values and process them simultaneously. As we demonstrate in this article, these existing individual or combined storage approaches for the wavelet coefficients of different measures can easily lead to suboptimal storage utilization, resulting in drastically reduced accuracy for approximate query answers. To address this problem, in this work, we introduce the notion of an extended wavelet coefficient as a flexible, efficient storage method for wavelet coefficients over multimeasure data. We also propose novel algorithms for constructing effective (optimal or near-optimal) extended wavelet-coefficient synopses under a given storage constraint, for both sum-squared error and relative-error norms. Experimental results with both real-life and synthetic datasets validate our approach, demonstrating that our techniques consistently obtain significant gains in approximation accuracy compared to existing solutions. Categories and Subject Descriptors: H.2.4 [Database Management]: Systems--Query processing General Terms: Algorithms, Experimentation, Performance Additional Key Words and Phrases: Wavelets, data synopses, approximate query processing ACM Reference Format: Deligiannakis, A., Garofalakis, M., and Roussopoulos, N. 2007. Extended wavelets for multiple measures. ACM Trans. Datab. Syst. 32, 2, Article 10 (June 2007), 52 pages. DOI = 10.1145/ 1242524.1242527 http://doi.acm.org/10.1145/1242524.1242527