11 result(s)
Page Size: 10, 20, 50
Export: bibtex, xml, json, csv
Order by:

CNR Author operator: and / or
Typology operator: and / or
Language operator: and / or
Date operator: and / or
Rights operator: and / or
2001 Journal article Restricted

Density parameter estimation of Skewed alfa-Stable distributions
Kuruoglu E. E.
Over the last few years, there has been a great interest in -stable distributions for modeling impulsive data. As a critical step in modeling with -stable distributions, the problem of estimating the parameters of stable distributions have been addressed by several works in the literature. However, many of these works consider only the special case of symmetric stable random variables. This is an important restriction, however, since most real-life signals are skewed. The existing techniques on estimating skewed distribution parameters are either computationally too expensive, require lookup tables, or have poor convergence properties. In this paper, we introduce three novel classes of estimators for the parameters of general stable distributions, which are generalizations of the methods previously suggested for parameter estimation of symmetric stable distributions. These estimators exploit expressions we develop for fractional lower order, negative order, and logarithmic moments and tail statistics. We also introduce simple transformations that allow one to use existing symmetric stable parameter estimation techniques. Techniques suggested in this paper provide the only closed-form solutions we are aware of for parameters that may be efficiently computed. Simulation results show that at least one of our new estimators has better performance than the existing techniques over most of the parameter space. Furthermore, our techniques require substantially less computation.Source: IEEE transactions on signal processing 49 (2001): 2192–2201.

See at: CNR ExploRA Restricted


2001 Journal article Open Access OPEN

Modelling SAR imaging of urban areas
Kuruoglu E. E., Zerubia J.
Satellite imagery has found vast applications in a wide spectrum of areas including agriculture (eg detection of crop types), urbanization (tracking the development of urban areas), cartography (eg detection of rivers, road networks), warfare (eg detection of targets, surveying), etc. This heavy demand on satellite imagery applications lead to the development of imaging systems that are alternative to optical imagery. In particular, synthetic aperture radar (SAR) imagery in the last two decades has become increasingly popular as some of its properties are favorable to optical imagery. SAR imagery can operate regardless of weather conditions and SAR image resolution is independent of sensor heighSource: ERCIM news 46 (2001): 68–69.

See at: CNR ExploRA Open Access


2001 Conference article Open Access OPEN

Modelling images with alpha-stable textures
Kuruoglu E. E., Zerubia J.
In this paper we present an alternative to Gaussian texture generated with alpha stable innovation.The advantage alpha stable textures present over Gaussian textures are twofold: they can model both textures exhibiting row impulsive ch`aracteristic and those that exhibit unsymmetrical characteristic.Source: PSIP'2001 Physics in Signal and Image Processing, pp. 5–9, Marseille, France, 23-24 January 2001

See at: CNR ExploRA Open Access


2001 Report Unknown

Blind separation of time-correlated sources from noisy data
Tonazzini A., Bedini L., Kuruoglu E. E., Salerno E.
An abstract is not available.Source: ISTI Technical reports, 2001

See at: CNR ExploRA


2001 Report Open Access OPEN

Density parameter estimation of skewed alfa-stable distributions
Kuruoglu E. E.
Over the last few years, there has been a great interest in alpha-stable distributions for modelling impulsive data. As a critical step in modelling with alpha-stable distributions, the problem of estimating the parameters of stable distributions have been addressed by several works in the literature. However, many of these works consider only the special case of symmetric stable random variables. This is an important restriction though, since most real life signals are skewed. The existing techniques on estimating skewed distribution parameters are either computationally too expensive, require lookup tables or have poor convergence properties. In this paper, we introduce three novel classes of estimators for the parameters of general stable distributions, which are generalisations of methods previously suggested for parameter estimation with symmetric stable distributions. These estimators exploit expressions we develop for fractional lower order, negative order and logarithmic moments and tail statistics. We also introduce simple transformations which allow one to use existing symmetric stable parameter estimation techniques. Techniques suggested in this paper provide the only closed form solutions we are aware of for parameters which may be efficiently computed. Simulation results show that at least one of our new estimators has better performance than the existing techniques over most of the parameter space. Furthermore our techniques require substantially less computation.Source: ISTI Technical reports, pp.1–30, 2001

See at: CNR ExploRA Open Access


2001 Report Open Access OPEN

Document image retrieval without OCRing using a video scanning system
Kuruoglu E. E., Vern Tan T.
We propose a technique for efficient document retrieval from digital libraries containing document images which are compressed with token based compression. The technique we propose uses the layout information supplied by the relative positions of the character tokens on the page of a 'query' paper document to retrieve the original document in the image database. The query image is captured from a paper document by a multimedia system composed of a PC and a video scanning tool. This technique avoids OCRing the query document and the documents in the database; moreover avoidsdecompressing the documents in the database compressed with token based compression, therefore achieving important time and computational gains. The technique provides one with the capability of retrieving the original document stored in a digital library using part of a previously produced paper copySource: ISTI Technical reports, pp.1–9, 2001

See at: CNR ExploRA Open Access


2001 Report Open Access OPEN

Independent factor analysis for component separation from planck channel maps
Kuruoglu E. E., Bedini L., Paratore M. T., Salerno E., Tonazzini A.
A microwave sky map results from a combination of signals from various astrophysical sources, such as cosmic microwave background radiation, synchrotron radiation and galactic dust radiation. To derive information about these sources, one needs to separate them from the measured maps on different frequency channels. This task is made difficult by our insufficient knowledge of the weights to be given to the individual signals at different frequencies. Recent work on the problem led to only limited success due to ignoring the noise and to the lack of a suitable statistical model for the sources. In this paper, we derive the statistical distribution of some source realizations, and check the appropriateness of a Gaussian mixture model for them. A source separation technique, namely independent factor analysis, had been suggested recently in the literature for Gaussian mixture sources in the presence of noise. This technique employs a three layered neural network architecture which allows a simple, hierarchical treatment of the problem. We modify the algorithm proposed in the literature to accommodate for space-varying noise and test its performance on simulated astrophysical maps. We also compare the performances of the expectation-maximization and the simulated annealing learning algorithm in the estimation of the mixing parameters. The simulation results demonstrate the success of the independent factor analysis approach with simulated-annealing learning, which proves better than the expectation-maximization learning especially for higher noise levels and when the independence of the performance from starting points is considered.Source: ISTI Technical reports, pp.1–26, 2001

See at: CNR ExploRA Open Access


2001 Report Open Access OPEN

Nonlinear least lp-norm filters for nonlinear autoregressive alpha-stable processes
Kuruoglu E. E.
The alpha-stable distribution family has received great interest recently, due to its ability to succesfully model impulsive data. alpha-stable distributions have found applications in areas such as radar signal processing, audio restoration, financial time series modelling and image processing. Various work on linear parametric models with alpha-stable innovations have been reported in the literature. However, some recent work has demonstrated that linear models are not in general adequate to capture all characteristics of heavy-tailed data. Moreover, it is known that the optimal minimum dispersion estimator for alpha-stable data is not necessarily linear. Therefore, in this paper, we suggest a shift in the interest to nonlinear parametric models for problems involving alpha-stable distributions. In particular, we study a simple yet analytic nonlinear random process model namely polynomial autoregressive alpha-stable processes. Polynomial autoregression and Volterra filtering have been successful models for some biomedical and seismic signals reflecting their underlying nonlinear generation mechanisms. In this paper, we employ alpha-stable processes instead of classical Gaussian distribution as innovation sequence and arrive at a model capable of describing unsymmetric as well as impulsive characteristics. We provide a number of novel adaptive and block type algorithms for the estimation of model parameters of this class of nonlinear processes efficiently. Simulation results on synthetic data demonstrate clearly the superiority of the novel algorithms to classical techniques. The paper concludes with a discussion of the application areas of the techniques developed in the paper, including impulsive noise suppression, nonlinear system identification, target tracking and nonlinear channel equalization.Source: ISTI Technical reports, pp.2–33, 2001

See at: CNR ExploRA Open Access


2001 Report Open Access OPEN

Time-frequency based detection in impulsive noise environments using alpha-stable noise models
Coates M., Kuruoglu E. E.
In various signal processing scenarios, the signal exhibits a nonstationary behaviour and is observed under non-Gaussian noise. Although one can find several work in the literature on dealing with nonstationary signals using time-frequency analysis and with non-Gaussian noise, these two problems have not been studied together. In this paper, we aim to address both nonstationarity and non-Gaussiannity in a unified manner. In particular, we develop test statistics for the detection of arbitrary nonstationary second order signals observed under impulsive noise. We model the impulsive noise with symmetric alpha-stable distributions which recently received interest in the signal processing community. We consider the problem of detecting such signals in the presence of unknown time-frequency and time-scale offsets and demonstrate that approximations to locally-optimal test statistics can be expressed in a manner conducive to efficient implementation using time-frequency or time-scale representations.Source: ISTI Technical reports, pp.1–16, 2001

See at: CNR ExploRA Open Access


2000 Patent Unknown

Method and apparatus for collaborative annotation of a document
Kuruoglu E. E., Taylor A. S., Seeger M., Taylor S. A.
Source: US7346841, Internazionale, 2008

See at: CNR ExploRA


2000 Conference article Restricted

Document image retrieval without OCRing using a video scanning system
Kuruoglu E. E., Tan V.
In this paper, we propose a technique for efficient document retrieval from digital libraries containing document images which are token based compressed. The query image is captured from a paper document by the video scanning tool of a multimedia system. The technique we propose uses the layout information supplied by the relative positions of the character tokens on the page of a "query" paper document to retrieve the original document in the image database. This technique avoids OCRing the query document and the documents in the database; moreover avoids decompressing the token based compressed documents in the database, therefore achieving important time and computational gains.Source: ACM Multimedia 2000 Workshops, pp. 233–236, Los Angeles, USA, 4 November 2000

See at: CNR ExploRA Restricted