2005
Journal article
Restricted
Image denoising using bivariate alpha-stable distributions in the complex wavelet domain
Achim A, Kuruoglu EeRecently, the dual-tree complex wavelet transform has been proposed as a novel analysis tool featuring near shift-invariance and improved directional selectivity compared to the standard wavelet transform. Within this framework, we describe a novel technique for removing noise from digital images. We design a bivariate maximum a posteriori (MAP) estimator, which relies on the family of isotropic alpha-stable distributions. Using this relatively new statistical model we are able to better capture the heavy-tailed nature of the data as well as the interscale dependencies of wavelet coefficients. We test our algorithm for the Cauchy case, in comparison with several recently published methods. The simulation results show that our proposed technique achieves state-of-the-art performance in terms of root mean squared error.Source: IEEE SIGNAL PROCESSING LETTERS, vol. 12 (issue 1), pp. 17-20
DOI: 10.1109/lsp.2004.839692Metrics:
See at:
IEEE Signal Processing Letters
| CNR IRIS
| ieeexplore.ieee.org
| CNR IRIS
2005
Journal article
Open Access
Separation of correlated astrophysical sources using multiple-lag data covariance matrices
Bedini L, Herranz D, Salerno E, Baccigalupi C, Kuruoglu Ee, Tonazzini AThis paper proposes a new strategy to separate astrophysical sources that are mutually correlated. This strategy is based on second order statistics and exploits prior information about the possible structure of the mixing matrix. Unlike ICA blind separation approaches, where the sources are assumed mutually independent and no prior knowledge is assumed about the mixing matrix, our strategy allows the independence assumption to be relaxed and performs the separation of even significantly correlated sources. Besides the mixing matrix, our strategy is also capable to evaluate the source covariance functions at several lags. Moreover, once the mixing parameters have been identified, a simple deconvolution can be used to estimate the probability density functions of the source processes. To benchmark our algorithm, we used a database that simulates the one expected from the instruments that will operate onboard ESA's Planck Surveyor Satellite to measure the CMB anisotropies all over the celestial sphere.Source: EURASIP JOURNAL ON APPLIED SIGNAL PROCESSING, vol. 2005 (issue 15), pp. 2400-2412
DOI: 10.1155/asp.2005.2400DOI: 10.48550/arxiv.astro-ph/0407108Metrics:
See at:
arXiv.org e-Print Archive
| EURASIP Journal on Advances in Signal Processing
| EURASIP Journal on Advances in Signal Processing
| EURASIP Journal on Advances in Signal Processing
| doi.org
| CNR IRIS
| CNR IRIS
2005
Journal article
Open Access
Editorial activity - Applications of signal processing in astrophysics and cosmology
Kuruoglu Ee, Baccigalupi CWe live in an epoch where the frontiers of our investigation and comprehension of fundamental physics depend largely on the light coming from the sky, that is, on the study of galactic and extra-galactic radiation. Watching the sky, in principle, we have access to the highest energies conceivable, generated by the laws of nature in extreme conditions, such as nearby black holes or even close to the origin of the universe itself. For example, in the microwave band, the extra-Galactic radiation is dominated by a markedly isotropic component, obeying a black body spectrum characterized by a temperature of about 2.726 Kelvin. That is the relic of the Big Bang, originated just 300 000 years after the initial starting point of the universe. This radiation, namely the cosmic microwave background (CMB) radiation, today is the most important observable we have to access the mysterious physics of the Big Bang itself. The latter is telling us about the unknown fundamental interactions and particles, the physics of spacetime, and the nature of quantum gravity, and represents the only way to address those issues in physics today. Electronics hardware technology has reached in these very recent years the capability to study the tiniest details of the CMB, carrying the image of the primordial stage of cosmic geometry, structure, and composition. Such a fantastic challenge is ongoing in this verymoment, while several CMB detectors are operating and advanced probes are being designed for the forthcoming decades. Many breakthroughs in physics are made possible by the use of the most advanced data analysis techniques. The present datasets obtained in astrophysical and cosmological observations are huge, and cover the entire electromagnetic spectrum, dealing with very different processes, from gamma and X-rays of the high-energy astrophysics of compact stars or black holes, to the microwave and infrared emission from the whole large-scale universe. This variety of the observational techniques and signals to deal with represents a formidable challenge for signal processing.We need state-ofthe-art techniques that can analyse, summarise, and extract the necessary information from this ocean of data. To continue with the example above, the microwave sky is dominated by the CMB radiation, but several processes contribute to the total emission, coming for instance fromall the processes occurring along the line of sight, such as the emission from other galaxies or clusters of those, as well as from the diffuse gas in our own Galaxy. Each of these processes are most relevant in different contexts in astrophysics and cosmology. Recently, the astrophysics field has benefited a great deal from the rich research work going on source separation in the signal processing field. Source separation aims at the recovery of the various different components fromthemultiband observations exploiting the differences between them, induced by their independent physical origins. Despite the mutual interest, the two disciplines suffer from lack of a common publication ground, implying that the results produced in one of them are not immediately visible in the other. The aim of the present issue is to provide a unified platform that would strengthen the bridge between signal processing and astrophysics and cosmology and enable the sharing of information. We would like to provide astrophysicists and cosmologists with a spectrum of the most advanced signal processing techniques and the signal processing community an exposure to various vital real problems in analysing astrophysics data that await solution. Finally, our aim is to provide a reference for present and future literature, in the widest possible context, accounting for various applications and algorithms proposed. Indeed, as the reader may see, the topics we collected range from solar physics, thus on the scale of stars, to the reconstructuon of the most ambitious signal from the Big Bang, with the reconstruction of the CMB pattern on all sky. The methods presented in the issue range from transform domain analysis of such wavelets to data mining techniques.Source: EURASIP JOURNAL ON APPLIED SIGNAL PROCESSING, vol. 15, pp. 2397-2399
See at:
CNR IRIS
| www.hindawi.com
| CNR IRIS
2005
Journal article
Restricted
A two-dimensional wavelet-based approach
Bozzi E, Cavaccini G, Chimenti M, Di Bono Mg, Salvetti OAn image processing procedure is proposed to detect porosity defects in composite materials, analyzing C-scan images obtained by ultrasound inspection techniques. An image described by a set of features is analyzed in order to evaluate its similarity with a reference set. A 2D wavelet transform is applied to the input image and then a feature extraction based on statistics of the detailed images produced by the transform itself is performed. The principal component analysis technique (PCA) is then applied in order to map input features into an output plane maximizing data variance. Finally the image is classified considering the distance between points in the PCA plane. This procedure is also applied for the analysis of a single image. Preliminary results on simulation images and real C-scan maps show that the procedure is able to detect defects.Source: PATTERN RECOGNITION AND IMAGE ANALYSIS, vol. 15-2, pp. 516-519
See at:
CNR IRIS
| CNR IRIS
2005
Journal article
Restricted
Automatic recognition and classification of cerebral microemboli in ultrasound images
Colantonio S, Salvetti O, Sartucci FThe aim of this work was the definition of a method devoted to the automated recognition of different composition of cerebral microemboli. The developed diagnostic procedure made use of a feature-based analysis of ultrasonographic images containing the characteristic microembolic signals. The images were acquired with a transcranial Doppler and classified using a hierarchical neural network. The proposed procedure was tested on clinical cases selected by expert neurologists for their relevance, and experimental results showed its reliability.Source: PATTERN RECOGNITION AND IMAGE ANALYSIS, vol. 15-2 (issue 2), pp. 532-535
See at:
CNR IRIS
| CNR IRIS
2005
Journal article
Restricted
Integration of two approaches to medical image analysis for diagnostic purposes
Di Bona S, Gurevich I, Koryabkina I, Nefyodov A, Salvetti OThis paper presents the results of the research activity performed in the field of medical image analysis within a joint study between the Institute of Information Science and Technologies of the Italian National Research Council (ISTI-CNR) and the Scientific Council 'Cybernetics' of the Russian Academy of Sciences (SCC-RAS). The studies carried out concern the analysis and classification of neuro (ISTI-CNR) and hematological (SCC-RAS) images. The comparison and integration of the approaches adopted by the two research groups have been fostered as an important activity to mutually improve the significant results obtained up to now by both ISTI-CNR and SCC-RAS in the field of medical imaging.Source: PATTERN RECOGNITION AND IMAGE ANALYSIS, vol. 15-2 (issue 2), pp. 539-542
See at:
CNR IRIS
| CNR IRIS
| puma.isti.cnr.it
2005
Journal article
Restricted
Processing multimedia biomedical information for disease evolution monitoring
Colantonio S, Di Bono Mg, Pieri GA methodology for the automatic monitoring of diseases using biomedical data in multimedia is proposed. We adopt a computational intelligence approach mainly based on a multilevel neural network architecture. This approach has been employed in applications for neurosignal and image categorisation.Source: ERCIM NEWS, vol. 60, pp. 28-29
See at:
CNR IRIS
| CNR IRIS
2005
Journal article
Restricted
Object detection and tracking in an open and free environment with a moving camera
Pieri G, Benvenuti M, Carnier E, Salvetti OThe task of detection and tracking of a moving object is addressed. An algorithm has been devel- oped which performs this task for monitoring and surveillance purposes. Prediction is also implemented in the algorithm to resolve the events of occlusion or masking, and also to increase the normal tracking performance. Real-time implementation generates deformation in the target appearance, and then a shape database is also used to improve losing target situation. A prototypical system has been developed that makes use of a moving camera located on a robotized system. A case study is presented on animal tracking in infrared live video.Source: PATTERN RECOGNITION AND IMAGE ANALYSIS, vol. 15-2 (issue 2), pp. 283-286
See at:
CNR IRIS
| CNR IRIS
2005
Journal article
Restricted
Representation and communication of multimedia data and metadata
Colantonio S, Di Bono Mg, Martinelli M, Pieri G, Salvetti OIn recent years the increasing role of Multimedia (MM) data, in the form of still pictures, graphics, 3D models, audio, speech, video or their combination (eg MM presentations), in the real world, has lead to a demand for better procedures for the automatic generation and extraction of both low level and semantic features from multi-source data in order to enhance their potential for computational interpretation and processing.Source: ERCIM NEWS, vol. 62, pp. 27-30
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Restricted
A methodological approach to the study of periodically deforming anatomical structures
Colantonio S, Moroni D, Salvetti OWe present a methodology, based on neural paradigms, suitable to analyze periodically deforming anatomical structures and recognize their state. Anatomical structures, considered as 'multimedia objects', are defined as organized sets of images and signals, acquired from multiple sources. These sets are combined and processed using dedicated Artificial Neural Networks to obtain a 3D reconstruction of the object, at different times of its dynamic evolution (deformation cycle). In order to reduce acquisition errors, we consider also an inter-cycle registration of the volumes, resulting in the most likely 3D reconstruction. Morphological and functional characteristics are then associated to each element of the reconstructed volume and processed to discriminate different object states. The developed methodology has been applied to analyze cardiac dynamics and, in particular, to identify physio-pathological states of the left ventricle.
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Unknown
A technique to optimize nonuniformly spaced arrays with low sidelobe level by using a genetic algorithm
Monorchio A., Genovesi S., Serra U., Brizzi A., Manara G.A Genetic Algorithm procedure for synthesizing the radiation pattern of nonuniformly spaced linear arrays with low side lobe level is presented. Some selected preliminary results are shown to validate the effectiveness and the reliability of the proposed approach.Source: 2005 IEEE AP-SInternational Symposium and USNC/URSI National Radio Science, Washington DC, USA, 3-8 July 2005
See at:
CNR ExploRA
2005
Conference article
Metadata Only Access
See at:
CNR IRIS
2005
Conference article
Restricted
Bayesian MRF-based blind source separation of convolutive mixtures of images
Tonazzini A, Gerace IThis paper deals with the recovery of clean images from a set of their noisy convolutive mixtures. In practice, this problem can be seen as the one of simultaneously separating and restoring source images that have been first degraded by unknown filters, then summed up and added with noise. We approach this problem in the framework of Blind Source Separation (BSS), where the unknown filters, in our case FIR filters in the form of blur kernels, must be estimated jointly with the sources. Assuming the statistical independence of the source images, we adopt Bayesian estimation for all the unknowns, and exploit information about local correlation within the individual sources through the use of suitable Gibbs priors, accounting also for well-behaved edges in the images. We derive an algorithm for recovering the blur kernels that make the estimated sources fit the known properties of the original sources. The method is validated through numerical experiments in a simplified setting, which is however related to real application scenarios.
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Restricted
Bayesian separation of non-stationary mixtures of dependent gaussian sources
Gencaga D, Kuruoglu Ee, Ertuzun An this work, we propose a novel approach to perform Dependent Component Analysis (DCA). DCA can be thought as the separation of latent, dependent sources from their observed mixtures which is a more realistic model than Independent Component Analysis (ICA) where the sources are assumed to be independent. In general, the sources can be spatiotemporally dependent and the mixing system may be non-stationary. Here, we propose a DCA algorithm, that combines concepts of particle filters and Markov Chain Monte Carlo (MCMC) methods in order to separate non-stationary mixtures of spatially dependent Gaussian sources.
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Restricted
Disease evolution prognosis based on multi-source signals and image analysis
Colantonio S, Di Bono Mg, Salvetti OA methodology to approach the automatic monitoring and prognosis of diseases evolution is proposed. We define a multilevel system architecture capable to process multi-source biomedical data according to a coarse-to-fine paradigm. An application regarding neuro-signals and image categorization is also considered as a case study. The proposed methodology even preliminary has shown to be a possible approach to prognosis activity, mainly if suitably integrated into a hybrid system for medical decision support.
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Restricted
Equipment and procedures for microwave nondestructive evaluation of lapideous materials
Bozzi E, Chimenti M, Genovesi S, Salerno E, Zucchelli AA procedure for in-field nondestructive evaluation of lapideous materials is described. A portable instrument has been developed to evaluate the average permittivity of the probed volume. This is based on measuring the resonance frequency of a microstrip patch sensor. Once this hardware is enabled to perform coherent measurements, the optimization of an edge-preserving energy functional can yield high-resolution permittivity range profiles.
See at:
CNR IRIS
| CNR IRIS
2005
Conference article
Restricted
Estimation of time-varying autoregressive symmetric alpha-stable processes using particle filters
Gencaga D, Kuruoglu Ee, Ertuzun AIn the last decade alpha-stable distributions have become a standard model for impulsive data. Especially the linear symmetric alpha-stable processes have found applications in various fields. When the process parameters are time-invariant, various techniques are available for estimation. However, time-invariance is an important restriction given that in many communications applications channels are time-varying. For such processes, we propose a relatively new technique, based on particle filters which obtained great success in tracking applications involving non-Gaussian signals and nonlinear systems. Since particle filtering is a sequential method, it enables us to track the time-varying autoregression coefficients of the alpha-stable processes. The method is tested both for abruptly and slowly changing autoregressive parameters of signals, where the driving noises are symmetric-alpha-stable processes and is observed to perform very well. Moreover, the method can easily be extended to skewed alpha-stable distributions.
See at:
CNR IRIS
| CNR IRIS