S. Fotopoulos | University of Patras (original) (raw)

Papers by S. Fotopoulos

IEE Proceedings G (Electronic Circuits and Systems), 1986

Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429), 2003

A multiresolution color image segmentation method is presented that incorporates the main princip... more A multiresolution color image segmentation method is presented that incorporates the main principles of region-based and cluster analysis approaches. A multiscale dissimilarity measure in the feature space is proposed that makes use of non-parametric cluster validity analysis and fuzzy C-Means clustering. Detected clusters are utilized to assign membership functions to the image regions. In addition, a graph theoretic merging algorithm is presented that uses the formulation of fuzzy similarity relations to produce the final segmentation results. The efficiency of the resulting scheme is also experimentally indicated.

Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205), 2001

Vision Systems: Segmentation and Pattern Recognition, 2007

Electroencephalography and Clinical Neurophysiology - Evoked Potentials, 1997

This technical note describes a robust version of moving averages, that enables reliable monitori... more This technical note describes a robust version of moving averages, that enables reliable monitoring of the evoked potential (EP) signals. A cluster analysis (CA) procedure is introduced to robustify the signal averaging (SA). It is implemented via a Hopfield neural network (HNN), which performs selection of the trials forming a cluster around the current state of the EP signal. The core of this cluster serves as an estimate of the instantaneous EP. The effectiveness of the method, indicated by application to real data, and its computation efficiency, due to the use of simple matrix operations, makes it very promising for clinical observations.

Pattern Recognition Letters, 2012

The last few years a growing research interest has aroused in the field of biometrics,concerning ... more The last few years a growing research interest has aroused in the field of biometrics,concerning the use of brain dependent characteristicsgenerally known as behavioral features. Human eyes, often referred as the gates to the soul, can possibly comprise a rich source of idiosyncratic information which may be used for the recognition of an individual's identity. In this paper an innovative experiment and a novel processing approach for the human eye movements is implemented, ultimately aiming at the biometric segregation of individual persons. In our experiment, the subjects observe face images while their eye movements are being monitored, providing information about each participant's attention spots. The implemented method treats eye trajectories as 2-D distributions of points on the image plane. The efficiency of graph objects in the representation of structural information motivated us on the utilization of a non-parametric multivariate graph-based measure for the comparison ofeye movement signals, yielding promising results at the task of identification according to behavioral characteristics of an individual.

Pattern Recognition, 2006

We address the problem of image similarity in the compressed domain, using a multivariate statist... more We address the problem of image similarity in the compressed domain, using a multivariate statistical test for comparing color distributions. Our approach is based on the multivariate Wald-Wolfowitz test, a nonparametric test that assesses the commonality between two different sets of multivariate observations. Using some pre-selected feature attributes, the similarity measure provides a comprehensive estimate of the match between different images based on graph theory and the notion of minimal spanning tree (MST). Feature extraction is directly provided from the JPEG discrete cosine transform (DCT) domain, without involving full decompression or inverse DCT. Based on the zigzag scheme, a novel selection technique is introduced that guarantees image's enhanced invariance to geometric transformations. To demonstrate the performance of the proposed method, the application on a diverse collection of images has been systematically studied in a query-by-example image retrieval task. Experimental results show that a powerful measure of similarity between compressed images can emerge from the statistical comparison of their pattern representations.

Multimedia Tools and Applications, 2009

The paper presents an automatic video summarization technique based on graph theory methodology a... more The paper presents an automatic video summarization technique based on graph theory methodology and the dominant sets clustering algorithm. The large size of the video data set is handled by exploiting the connectivity information of prototype frames that are extracted from a down-sampled version of the original video sequence. The connectivity information for the prototypes which is obtained from the whole set of data improves video representation and reveals its structure. Automatic selection of the optimal number of clusters and hereafter keyframes is accomplished at a next step through the dominant set clustering algorithm. The method is free of user-specified modeling parameters and is evaluated in terms of several metrics that quantify its content representational ability. Comparison of the proposed summarization technique to the Open Video storyboard, the Adaptive clustering algorithm and the Delaunay clustering approach, is provided.

IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 2005

IEEE Transactions on Biomedical Engineering, 1995

The ensemble average of Pattern Shift Visual Evoked Potentials (PSVEP) signals is seriously affec... more The ensemble average of Pattern Shift Visual Evoked Potentials (PSVEP) signals is seriously affected by random latency variations encountered in each individual sweep which is modeled as a continuous signal with linear segments and well-shaped triangular peaks. This effect is causing the smoothed peaks of the averaged PSVEP waveforms. It is our objective to restore the degraded peaks and provide accurate information about their exact location. The method used is based on nonlinear filtering of the FIR-Median Hybrid (FMH) type and is recommended as a postfiltering process to the well-known averaging methods of recovering PSVEP signals from noise by time-locking to stimuli. The new technique, tested in signals from clinical observations, has proven very promising.

Computer Vision and Image Understanding, 2006

... have experimented with different color spaces (eg, RGB, CIE-Lab), various color descriptors (... more ... have experimented with different color spaces (eg, RGB, CIE-Lab), various color descriptors (eg, color ... overcome the above limitation, an attempt was made recently [14] to combine the benefits ... images (A) and dissimilar images (B), based on the k = 15 color prototypes shown in ...

In this work a color based, edge-region cooperative segmentation technique is presented. The segm... more In this work a color based, edge-region cooperative segmentation technique is presented. The segmentation process consists of two basic stages. In the first stage a novel fast vector edge detector based on the image density map is used, to solve the multidimensional boundary estimation problem. The watershed transform with some form of morphological prefiltering, is applied on the edge image to produce well defined regions. In the final stage a new vector merging criterion is used to control oversegmentation. Experimental results on natural images are presented. The segmentation procedure preserves color in every stage.

Pattern Recognition

The presence of Antinuclear Autoantibodies (ANA) in human serum is connected with several autoimm... more The presence of Antinuclear Autoantibodies (ANA) in human serum is connected with several autoimmune diseases. Indirect Immunofluorescence (IIF) imaging of human epithelial type-2 cells (HEp-2) is the dominant protocol used for the identification of ANA. However, due to limitations in the processes, several attempts have been made to automate the procedure of HEp-2 cell classification. In this work, we focus on the task of HEp-2 cell classification and we propose a novel method for local feature encoding that allows us to generalize the concept of residual encoding in sparse vectors. More specifically, our method hierarchically aggregates the residual of the feature vectors' sparse representation leading to a Vector of Hierarchically Aggregated Residuals (VHAR). Using SIFT descriptors computed on a dense grid and multiple scales, as well as considering spatial information our method achieves 78.0% classification and 82.3% mean class rate (MCR) on ICPR2012 and ICPR2014 HEp-2 cell contest datasets respectively. A novel method for encoding features' Sparse Representation residuals is proposed.Multiple levels of unsupervised learning are utilized.Local gradient descriptors are encoded into fixed length vectors.The method is evaluated on the task of HEp-2 classification.The proposed framework follows recent trends on feature encoding based on residuals.

2000 10th European Signal Processing Conference, Sep 1, 2000

A novel hierarchical clustering method is presented in this work. It operates as a part of a spli... more A novel hierarchical clustering method is presented in this work. It operates as a part of a split and merge segmentation scheme. The proposed technique incorporates the use of several color features to compare clusters in the RGB space and the flexibility of the fuzzy reasoning approach to accomplish satisfactory segmentation results. The boundary values of the fuzzy sets have been determined by means of a genetic algorithm optimization approach. The segmentation results evaluated subjectively and objectively were compared to a straightforward product cost function.

Electronics Letters, 2000

... Opt. Teclznol. Lett., 1995, 8, (6), pp. 282-287 ROY, &amp;amp;#x27;r., SARKAR, TK, DJ... more ... Opt. Teclznol. Lett., 1995, 8, (6), pp. 282-287 ROY, &amp;amp;#x27;r., SARKAR, TK, DJORDJEVIC, AR. aiidSALAZA, M : &amp;amp;#x27;A hybrid method solution of scattering by conducting cylinders (TM case)&amp;amp;#x27;, IEEE Trans. Microw. ... 97-104 ROY, T., SARKAR, TK, DJORD.IEVIC, AR. and SALAZA, M.: &amp;amp;#x27;A hybrid ...

International Journal of Psychophysiology, 1994

ABSTRACT In this paper, we introduce new weighted FIR median hybrid (FMH) filters consisting of c... more ABSTRACT In this paper, we introduce new weighted FIR median hybrid (FMH) filters consisting of combinations of averaging and ramp predicting FIR substructures, the input signal and the weighted median operation. We analyze two weighted FMH filter structures in detail. Due to the selection of subfilters the center weighted FMH (CWFMH) filter structure has low-pass characteristics and it preserves sinusoidal signals. It is shown that sinusoidal signals are root signals of the filter. An upper frequency bound for the preservation of sinusoids is derived. The subfilter weighted FMH (SWFMH) filter has the property of preserving edges while filtering high-frequency stationary signals. The characteristics of both filters are analyzed for sinusoidal and pulse shaped input signals. An efficient implementation structure for the weighted FMH filters with complexity independent of the filter length is shown. As examples, removal of artifacts and noise from the electroencephalogram (EEG) and electrooculogram (EOG) signals, respectively, is studied.