Human Machines Interaction HMI Laboratory

Kavála, Greece

Human Machines Interaction HMI Laboratory

Kavála, Greece

Time filter

Source Type

Papakostas G.A.,Human Machines Interaction HMI Laboratory | Savio A.,University of the Basque Country | Grana M.,University of the Basque Country | Kaburlasos V.G.,Human Machines Interaction HMI Laboratory
Neurocomputing | Year: 2015

We present a Computer Assisted Diagnosis (CAD) system for Alzheimer's disease (AD). The proposed CAD system employs MRI data features and applies a Lattice Computing (LC) scheme. To this end feature extraction methods are adopted from the literature, toward distinguishing healthy people from Alzheimer diseased ones. Computer assisted diagnosis is pursued by a k-NN classifier in the LC context by handling this task from two different perspectives. First, it performs dimensionality reduction over the high dimensional feature vectors and, second it classifies the subjects inside the lattice space by generating adaptively class boundaries. Computational experiments using a benchmark MRI dataset regarding AD patients demonstrate that the proposed classifier performs well comparatively to state-of-the-art classification models. © 2014 Elsevier B.V.


Papakostas G.A.,Human Machines Interaction HMI Laboratory | Koulouriotis D.E.,Democritus University of Thrace | Karakasis E.G.,Democritus University of Thrace | Tourassis V.D.,Democritus University of Thrace
Neurocomputing | Year: 2013

A novel descriptor able to improve the classification capabilities of a typical pattern recognition system is proposed in this paper. The introduced descriptor is derived by incorporating two efficient region descriptors, namely image moments and local binary patterns (LBP), commonly used in pattern recognition applications, in the last decades. The main idea behind this novel feature extraction methodology is the need of improved recognition capabilities, a goal achieved by the combinative use of these descriptors. This collaboration aims to make use of the major advantages each one presents, by simultaneously complementing each other, in order to elevate their weak points. In this way, the useful properties of the moments and moment invariants regarding their robustness to the noise presence, their global information coding mechanism and their invariant behaviour under scaling, translation and rotation conditions, along with the local nature of the LBP, are combined in a single concrete methodology. As a result a novel descriptor invariant to common geometric transformations of the described object, capable to encode its local characteristics, is formed and its classification capabilities are investigated through massive experimental scenarios. The experiments have shown the superiority of the introduced descriptor over the moment invariants, the LBP operator and other well-known from the literature descriptors such as HOG, HOG-LBP and LBP-HF. © 2012 Elsevier B.V.


Papadakis S.E.,Technological Educational Institute of Crete | Kaburlasos V.G.,Human Machines Interaction HMI Laboratory | Papakostas G.A.,Human Machines Interaction HMI Laboratory
World Scientific Proc. Series on Computer Engineering and Information Science 7; Uncertainty Modeling in Knowledge Engineering and Decision Making - Proceedings of the 10th International FLINS Conf. | Year: 2012

We deal with the problem of human facial expression recognition. An image is preprocessed for feature extraction using moment descriptors; then, it is represented in the product lattice (F1 100, ≤) of Intervals' Numbers (INs). Both learning and generalization are pursued in (F1 100, ≤) by a Fuzzy Lattice Reasoning (FLR) classifier based on an inclusion measure function σ: F1 100 x× F1 100 → [0, 1]. Preliminary, experiments on a benchmark set have been promising.


Kaburlasos V.G.,Human Machines Interaction HMI Laboratory | Kehagias A.,Aristotle University of Thessaloniki
IEEE Transactions on Fuzzy Systems | Year: 2014

A fuzzy inference system (FIS) typically implements a function f: ℝN: → T, where the domain set ℝ denotes the totally ordered set of real numbers, whereas the range set T may be either T = ℝM (i.e., FIS regressor) or T may be a set of labels (i.e., FIS classifier), etc. This study considers the complete lattice (F,≺) of Type-1 Intervals' Numbers (INs), where an IN F can be interpreted as either a possibility distribution or a probability distribution. In particular, this study concerns the matching degree (or satisfaction degree, or firing degree) part of an FIS. Based on an inclusion measure function σ : F × F → [0,1] we extend the traditional FIS design toward implementing a function f: FN → T with the following advantages: 1) accommodation of granular inputs; 2) employment of sparse rules; and 3) introduction of tunable (global, rather than solely local) nonlinearities as explained in the manuscript. New theorems establish that an inclusion measure σ is widely (though implicitly) used by traditional FISs typically with trivial (i.e., point) input vectors. A preliminary industrial application demonstrates the advantages of our proposed schemes. Far-reaching extensions of FISs are also discussed. © 2014 IEEE.


Papakostas G.A.,Human Machines Interaction HMI Laboratory | Hatzimichailidis A.G.,Human Machines Interaction HMI Laboratory | Kaburlasos V.G.,Human Machines Interaction HMI Laboratory
Pattern Recognition Letters | Year: 2013

A detailed analysis of the distance and similarity measures for intuitionistic fuzzy sets proposed in the past is presented in this paper. This study aims to highlight the main theoretical and computational properties of the measures under study, while the relationships between them are also investigated. Along with the literature review, a comparison of the analyzed distance and similarity measures from a pattern recognition point of view in three different classification cases is also presented. Initially, some artificial counter-intuitive recognition cases are considered, while in a second phase real data from medical and well known pattern recognition benchmark problems are used to examine the discrimination abilities of the studied measures. Moreover, all the measures are applied in a face recognition problem for the first time and useful conclusions are drawn regarding the accuracy and confidence of the recognition results. Finally, the measures' suitability and their drawbacks that make the development of more robust and efficient measures' a still open issue are discussed. © 2013 Elsevier B.V. All rights reserved.


Karakasis E.G.,Democritus University of Thrace | Papakostas G.A.,Human Machines Interaction HMI Laboratory | Koulouriotis D.E.,Democritus University of Thrace | Tourassis V.D.,Democritus University of Thrace
Pattern Recognition | Year: 2013

In this work we introduce a generalized expression of the weighted dual Hahn moment invariants up to any order and for any value of their parameters. In order for the proposed invariants to be formed, the weighted dual Hahn moments (up to any order and for any value of their parameters) are expressed as a linear combination of geometric ones. For this reason a formula expressing the nth degree dual Hahn polynomial, for any value of its parameters, as a linear combination of monomials (cr·xr), is proved. In addition, a recurrent relation for the fast computation of the aforementioned monomials coefficients (cr) is also given. Moreover, normalization aspects of the generalized weighted dual Hahn moment invariants are discussed, while a modification of them is proposed in order to avoid their numerical instabilities. Finally, experimental results and classification scenarios, including datasets of natural scenes, evaluate the proposed methodology. © 2013 Elsevier Ltd. All rights reserved.


Kaburlasos V.G.,Human Machines Interaction HMI Laboratory | Papadakis S.E.,Technological Educational Institute of Crete | Papadakis S.E.,HMI Laboratory | Papakostas G.A.,Human Machines Interaction HMI Laboratory
IEEE Transactions on Neural Networks and Learning Systems | Year: 2013

This paper proposes a fundamentally novel extension, namely, flrFAM, of the fuzzy ARTMAP (FAM) neural classifier for incremental real-time learning and generalization based on fuzzy lattice reasoning techniques. FAM is enhanced first by a parameter optimization training (sub)phase, and then by a capacity to process partially ordered (non)numeric data including information granules. The interest here focuses on intervals' numbers (INs) data, where an IN represents a distribution of data samples. We describe the proposed flrFAM classifier as a fuzzy neural network that can induce descriptive as well as flexible (i.e., tunable) decision-making knowledge (rules) from the data. We demonstrate the capacity of the flrFAM classifier for human facial expression recognition on benchmark datasets. The novel feature extraction as well as knowledge- representation is based on orthogonal moments. The reported experimental results compare well with the results by alternative classifiers from the literature. The far-reaching potential of fuzzy lattice reasoning in human-machine interaction applications is discussed. © 2013 IEEE.


Tsougenis E.D.,Democritus University of Thrace | Papakostas G.A.,Human Machines Interaction HMI Laboratory | Koulouriotis D.E.,Democritus University of Thrace
Multimedia Tools and Applications | Year: 2015

The use of image moments as host coefficients constitutes one of the hot topics in image watermarking field due to their robust behavior. Recenlty, a new approach called separable moments (SMs) has been introduced representing an image as combinations of different orthogonal polynomials that generate a series of new moment families. The scope of the present work is to introduce the specific transformations to the image watermarking field by evaluating their security capability under a wide range of common signal processing and geometric attacks. Furthermore, their ability of carrying large binary watermark messages is also examined. The performance of the proposed moment families is evaluated by a comparison to the original moments and a state-of-the-art method. The experimental results justified that a number of the studied transformations outperforms the benchmark method and occasionally the original moment families. Moreover, specific separable moment families are free of instabilities to the higher order coefficients where the extra watermark information is carried. A significant conclusion lies on the adoption of properties (locality, stability) between the generated separable moment families that lead to the enhancement of the basic watermarking requirements (robustness, imperceptibility and capacity) of the proposed watermarking method. The present work justifies that discrete orthogonal SMs constitute a new attractive transformation for the image moment-based watermarking field. © 2013, Springer Science+Business Media New York.


Tsougenis E.D.,Democritus University of Thrace | Papakostas G.A.,Human Machines Interaction HMI Laboratory | Koulouriotis D.E.,Democritus University of Thrace | Tourassis V.D.,Democritus University of Thrace
Optics and Laser Technology | Year: 2013

A successful image watermarking method is identified by the high performance in a number of basic requirements such as robustness, imperceptibility, capacity and complexity. Enhancement could be achieved through an adaptive process that handles individually the embedded information to each coefficient. The specific need for adaptivity is justified through this work by a set of experiments applied to the traditional moment families (Zernike, Pseudo-Zernike, Tchebichef), where more optimum results are produced. The extensive study of Polar Harmonic Transforms' (PHTs) significance parameters (order, magnitude) along with the use of a generalized embedding strength calculation process, easily applied to circularly orthogonal transformations, leads to a promising solution of the adaptivity issue. Experimental results justify that the proposed image watermarking scheme clearly outperforms the compared methods in terms of robustness, capacity and complexity and promotes the traditional schemes to a next generation of moment-based image watermarking. © 2013 Elsevier Ltd. All rights reserved.


Papadakis S.E.,Technological Educational Institute of Crete | Papadakis S.E.,Human Machines Interaction HMI Laboratory | Kaburlasos V.G.,Human Machines Interaction HMI Laboratory | Papakostas G.A.,Human Machines Interaction HMI Laboratory
Journal of Multiple-Valued Logic and Soft Computing | Year: 2014

We deal with the problem of human facial expression recognition from digital images. A digital image is preprocessed for feature extraction using moment descriptors; then, it is represented in the product lattice (F 100,≥) of Intervals'Numbers (INs). Learning as well as generalization are carried out in space (F100,≥) by two different Fuzzy Lattice Reasoning (FLR) classifiers based on an inclusion measure function σ : F100 × F100 → [0, 1].We pursue both a stochastic optimization and a parallel implementation of the proposed techniques. Comparative experimental results on three benchmark data sets demonstrate a superior performance of the proposed FLR classification schemes. © 2014 Old City Publishing, Inc.

Loading Human Machines Interaction HMI Laboratory collaborators
Loading Human Machines Interaction HMI Laboratory collaborators