Nikolas P. Galatsanos

University of Patras, Rhion, West Greece, Greece

Are you Nikolas P. Galatsanos?

Claim your profile

Publications (84)48.14 Total impact

  • Source
    Ashraf A. Tahat, Nikolas P. Galatsanos
    J. Electrical and Computer Engineering. 01/2010; 2010.
  • [Show abstract] [Hide abstract]
    ABSTRACT: A relevance feedback (RF) approach for content based image retrieval (CBIR) is proposed, which combines Support Vector Machines (SVMs) with Gaussian Mixture (GM) models. Specifically, it constructs GM models of the image features distribution to describe the image content and trains an SVM classifier to distinguish between the relevant and irrelevant images according to the preferences of the user. The method is based on distance measures between probability density functions (pdfs), which can be computed in closed form for GM models. In particular, these distance measures are used to define a new SVM kernel function expressing the similarity between the corresponding images modeled as GMs. Using this kernel function and the user provided feedback examples, an SVM classifier is trained in each RF round, resulting in an updated ranking of the database images. Numerical experiments are presented that demonstrate the merits of the proposed relevance feedback methodology and the advantages of using GMs for image modeling in the RF framework.
    Artificial Intelligence Applications and Innovations III, Proceedings of the 5TH IFIP Conference on Artificial Intelligence Applications and Innovations (AIAI'2009), April 23-25, 2009, Thessaloniki, Greece; 01/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A relevance feedback (RF) approach for content-based image retrieval (CBIR) is proposed, which is based on Support Vector Machines (SVMs) and uses a feature selection technique to reduce the dimensionality of the image feature space. Specifically, each image is described by a multidimensional vector combining color, texture and shape information. In each RF round, the positive and negative examples provided by the user are used to determine a relatively small number of the most important features for the corresponding classification task, via a feature selection methodology. After the feature selection has been performed, an SVM classifier is trained to distinguish between relevant and irrelevant images according to the preferences of the user, using the restriction of the user examples on the set of selected features. The trained classifier is subsequently used to provide an updated ranking of the database images represented in the space of the selected features. Numerical experiments are presented that demonstrate the merits of the proposed relevance feedback methodology.
    Artificial Neural Networks - ICANN 2009, 19th International Conference, Limassol, Cyprus, September 14-17, 2009, Proceedings, Part I; 01/2009
  • Source
    Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    IEEE Transactions on Image Processing 01/2009; 18:753-764. · 3.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We propose an approach to analyzing functional neuroimages in which 1) regions of neuronal activation are described by a superposition of spatial kernel functions, the parameters of which are estimated from the data and 2) the presence of activation is detected by means of a generalized likelihood ratio test (GLRT). Kernel methods have become a staple of modern machine learning. Herein, we show that these techniques show promise for neuroimage analysis. In an on-off design, we model the spatial activation pattern as a sum of an unknown number of kernel functions of unknown location, amplitude, and/or size. We employ two Bayesian methods of estimating the kernel functions. The first is a maximum a posteriori (MAP) estimation method based on a Reversible-Jump Markov-chain Monte-Carlo (RJMCMC) algorithm that searches for both the appropriate model complexity and parameter values. The second is a relevance vector machine (RVM), a kernel machine that is known to be effective in controlling model complexity (and thus discouraging overfitting). In each method, after estimating the activation pattern, we test for local activation using a GLRT. We evaluate the results using receiver operating characteristic (ROC) curves for simulated neuroimaging data and example results for real fMRI data. We find that, while RVM and RJMCMC both produce good results, RVM requires far less computation time, and thus appears to be the more promising of the two approaches.
    IEEE Transactions on Medical Imaging 01/2008; 26(12):1613-24. · 4.03 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper a relevance feedback (RF) approach for content based image retrieval (CBIR) is described and evaluated. The approach uses Gaussian mixture (GM) models of the image features and a query that is updated in a probabilistic manner. This update reflects the preferences of the user and is based on the models of both positive and negative feedback images. Retrieval is based on a recently proposed distance measure between probability density functions (pdfs), which can be computed in closed form for GM models. The proposed approach takes advantage of the form of this distance measure and updates it very efficiently based on the models of the user specified relevant and irrelevant images. For evaluation purposes, comparative experimental results are presented that demonstrate the merits of the proposed methodology.
    20th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2008), November 3-5, 2008, Dayton, Ohio, USA, Volume 1; 01/2008
  • Source
    IEEE Transactions on Image Processing 01/2008; 17:1795-1805. · 3.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a spatially constrained mixture model for the segmentation of MR brain images is presented. The novelty of this work is an edge-preserving smoothness prior which is imposed on the probabilities of the voxel labels. This prior incorporates a line process, which is modeled as a Bernoulli random variable, in order to preserve edges between tissues. The main difference with other, state of the art methods imposing priors, is that the constraint is imposed on the probabilities of the voxel labels and not onto the labels themselves. Inference of the proposed Bayesian model is obtained using variational methodology and the model parameters are computed in closed form. Numerical experiments are presented where the proposed model is favorably compared to state of the art brain segmentation methods as well as to a spatially varying Gaussian mixture model.
    Medical Image Computing and Computer-Assisted Intervention - MICCAI 2008, 11th International Conference, New York, NY, USA, September 6-10, 2008, Proceedings, Part I; 01/2008
  • Source
    Vasileios Chasanis, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we describe a system for video rushes summarization. The basic problems of rushes videos are three. First, the presence of useless frames such as colorbars, monochrome frames and frames containing clapboards. Second, the repetition of similar segments produced from multiple takes of the same scene and finally, the efficient representation of the original video in the video summary. In the method we proposed herein, the input video is segmented into shots. Then, colorbars and monochrome frames are removed by checking their edge direction histogram, whereas frames containing clapboards are removed by checking their SIFT descriptors. Next, an enhanced spectral clustering algorithm that both estimates the number of clusters and employs the fast global k-means algorithm in the clustering stage after the eigenvector computation of the similarity matrix is used to extract the key-frames of each shot, to efficiently represent shot content. Similar shots are clustered in one group by comparing their key-frames using a sequence alignment algorithm. Each group is represented from the shot with the largest duration and the final video summary is generated by concatenating frames around the key-frames of each shot. Experiments on TRECVID 2008 Test Data indicate that our method exhibits good performance.
    Proceedings of the 2nd ACM Workshop on Video Summarization, TVS 2008, Vancouver, British Columbia, Canada, October 31, 2008; 01/2008
  • Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: Recently, sparse kernel methods such as the Relevance Vector Machine (RVM) have become very popular for solving regression problems. The sparsity and performance of these methods depend on selecting an appropriate kernel function, which is typically achieved using a cross-validation procedure. In this paper we propose a modification to the incremental RVM learning method, that also learns the location and scale parameters of Gaussian kernels during model training. More specifically, in order to effectively model signals with different characteristics at various locations, we learn different parameter values for each kernel, resulting in a very flexible model. In order to avoid overfitting we use a sparsity enforcing prior that controls the effective number of parameters of the model. Finally, we apply the proposed method to one-dimensional and two-dimensional artificial signals, and evaluate its performance on two real-world datasets.
    Artificial Intelligence: Theories, Models and Applications, 5th Hellenic Conference on AI, SETN 2008, Syros, Greece, October 2-4, 2008. Proceedings; 01/2008
  • Source
    Konstantinos Blekas, Nikolas P. Galatsanos, Aristidis Likas
    Artificial Intelligence: Theories, Models and Applications, 5th Hellenic Conference on AI, SETN 2008, Syros, Greece, October 2-4, 2008. Proceedings; 01/2008
  • Antonis Mairgiotis, Nikolas P. Galatsanos, Yongyi Yang
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a new family of watermark detectors for additive watermarks in digital images. These detectors are based on a recently proposed hierarchical, two-level image model, which was found to be beneficial for image recovery problems. The top level of this model is defined to exploit the spatially varying local statistics of the image, while the bottom level is used to characterize the image variations along two principal directions. Based on this model, we derive a class of detectors for the additive watermark detection problem, which include a generalized likelihood ratio, Bayesian, and Rao test detectors. We also propose methods to estimate the necessary parameters for these detectors. Our numerical experiments demonstrate that these new detectors can lead to superior performance to several state-of-the-art detectors.
    IEEE Transactions on Information Forensics and Security. 01/2008; 3:29-37.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Advancements in the diagnosis and prognosis of brain tumor patients, and thus in their survival and quality of life, can be achieved using biomarkers that facilitate improved tumor typing. We introduce and implement a combinatorial metabolic and molecular approach that applies state-of-the-art, high-resolution magic angle spinning (HRMAS) proton (1H) MRS and gene transcriptome profiling to intact brain tumor biopsies, to identify unique biomarker profiles of brain tumors. Our results show that samples as small as 2 mg can be successfully processed, the HRMAS 1H MRS procedure does not result in mRNA degradation, and minute mRNA amounts yield high-quality genomic data. The MRS and genomic analyses demonstrate that CNS tumors have altered levels of specific 1H MRS metabolites that directly correspond to altered expression of Kennedy pathway genes; and exhibit rapid phospholipid turnover, which coincides with upregulation of cell proliferation genes. The data also suggest Sonic Hedgehog pathway (SHH) dysregulation may play a role in anaplastic ganglioglioma pathogenesis. That a strong correlation is seen between the HRMAS 1H MRS and genomic data cross-validates and further demonstrates the biological relevance of the MRS results. Our combined metabolic/molecular MRS/genomic approach provides insights into the biology of anaplastic ganglioglioma and a new potential tumor typing methodology that could aid neurologists and neurosurgeons to improve the diagnosis, treatment, and ongoing evaluation of brain tumor patients.
    International Journal of Molecular Medicine 09/2007; 20(2):199-208. · 1.96 Impact Factor
  • Christophoros Nikou, Nikolas P. Galatsanos, Aristidis Likas
    IEEE Transactions on Image Processing. 01/2007; 16:1121-1130.
  • Source
    Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: x n, tn)}N n=1. The weights are typically assigned those values that maximize the likelihood of the training set, however the training examples must be significantly more than the parameters in order to achieve good generalization performance. The RVM overcomes this limita- tion by following Bayesian principles and assuming prior knowledge for the model. Specifically, a suitable hierarchical prior distribution is assumed for the weights of
    International Journal of Artificial Intelligence Tools 01/2007; 16:967-979. · 0.25 Impact Factor
  • Source
    Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we present a new Bayesian model for the blind image deconvolution (BID) problem. The main novelties of this model are three. The first one is the use of a sparse kernel-based model for the point spread function (PSF) that allows estimation of both PSF shape and support. The second one is a robust distribution of the BID model errors and the third novelty is an image prior that preserves edges of the reconstructed image. Sparseness, robustness and preservation of edges is achieved by using priors that are based on the Student-t probability density function (pdf). The Variational methodology is used to solve the corresponding Bayesian model. Numerical experiments are presented that demonstrate the advantages of this model as compared to previous Gaussian based ones.
    Proceedings of the International Conference on Image Processing, ICIP 2007, September 16-19, 2007, San Antonio, Texas, USA; 01/2007
  • Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: ,In this paper we present a new Bayesian model for the blind image deconvolution (BID) problem. The main
    VISAPP 2007: Proceedings of the Second International Conference on Computer Vision Theory and Applications, Barcelona, Spain, March 8-11, 2007 - Volume Special Sessions; 01/2007
  • Brian C. Tom, Nikolas P. Galatsanos, Aggelos K. Katsaggelos
    [Show abstract] [Hide abstract]
    ABSTRACT: In this chapter the problem of reconstructing a high resolution image from multiple aliased and shifted by sub-pixel shifts low resolution images is considered. The low resolution images are possibly degraded by unknown blurs and their sub-pixel shifts are not known. This problem is described in the frequency and spatial domains. Algorithms for providing solutions to it are reviewed. In addition, two approaches are presented in detail for solving this low-to-high resolution problem. In the first of these two approaches registration and restoration is performed simultaneously using the expectation-maximization (EM) algorithm. The high resolution image is then reconstructed using regularized interpolation which is performed as a separate step. For this reason this approach is abbreviated as RR-I which corresponds to registration/restorationinterpolation. In the second of these approaches registration, restoration and interpolation are perfomed simultaneously using the EM algorithm. Therefore this approach is abbreviated as RRI which corresponds to registration/restoration/interpolation. Numerical experiments are presented that demonstrate the effectiveness of the two approaches.
    04/2006: pages 73-105;
  • Ahmad Abu-Naser, Nikolas P Galatsanos, Miles N Wernick
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigate the problem of detecting and localizing a known signal in a photon-limited image, where Poisson noise is the dominant source of image degradation. For this purpose we developed and evaluated three new algorithms. The first two are based on the impulse restoration (IR) principle and the third is based on the generalized likelihood ratio test (GLRT). In the IR approach, the problem is formulated as one of restoring a delta function at the location of the desired object. In the GLRT approach, which is a well-known variation on the optimal likelihood ratio test, the problem is formulated as a hypothesis testing problem, in which the unknown background intensity of the image and the intensity scale of the object are obtained by maximum-likelihood estimation. We used Monte Carlo simulations and localization receiver operating characteristic (LROC) curves to evaluate the proposed algorithms quantitatively. LROC curves demonstrate the ability of an algorithm to detect and locate objects in a scene correctly. Our simulations demonstrate that the GLRT approach is superior to all other tested algorithms.
    Journal of the Optical Society of America A 03/2006; 23(2):272-8. · 1.67 Impact Factor
  • Dimitris Tzikas, Aristidis Likas, Nikolas P. Galatsanos
    [Show abstract] [Hide abstract]
    ABSTRACT: The Relevance Vector Machine(RVM) is a widely accepted Bayesian model commonly used for regression and classification tasks. In this paper we propose a multikernel version of the RVM and present an alternative inference algorithm based on Fourier domain computation to solve this model for large scale problems, e.g. images. We then apply the proposed method to the object detection problem with promising results. where {`m(x)}M m=1 is a set of basis functions. Learning on such a model, is the process of estimating the weights {wm}M m=1 given a training set {(xn,tn)}N n=1. The weights are typically assigned those values that maximize the likelihood of the training set, however the training examples must be significantly more than the parameters in order to achieve good generalization performance. The RVM overcomes this limitation by following Bayesian principles and assuming prior knowledge for the model. Specifically, a suitable hierarchical prior distribution is assumed for the weights of the model, which has most probability mass con- centrated in sparse solutions, meaning that it forces most of the weights to be assigned to zero values (1). This results in pruning basis functions that are not suciently
    Advances in Artificial Intelligence, 4th Helenic Conference on AI, SETN 2006, Heraklion, Crete, Greece, May 18-20, 2006, Proceedings; 01/2006

Publication Stats

711 Citations
48.14 Total Impact Points

Institutions

  • 2008
    • University of Patras
      • Department of Electrical and Computer Engineering
      Rhion, West Greece, Greece
  • 2004–2007
    • University of Ioannina
      • • Τμήμα Πληροφορικής
      • • Laboratory of Computer Science
      Ioánnina, Ipeiros, Greece
    • Washington University in St. Louis
      San Luis, Missouri, United States
  • 1992–2006
    • Illinois Institute of Technology
      • Department of Electrical & Computer Engineering
      Chicago, IL, United States
  • 2002
    • Singapore-MIT Alliance
      Cambridge, Massachusetts, United States
  • 2000
    • Northeastern University
      Boston, Massachusetts, United States
  • 1994–1996
    • AT&T Labs
      Austin, Texas, United States
  • 1993–1995
    • Northwestern University
      • Department of Electrical Engineering and Computer Science
      Evanston, Illinois, United States