Computer and Automation Research Institute
Computer and Automation Research Institute
Kovacs A.,Pázmány Péter Catholic University |
Kovacs A.,Computer and Automation Research Institute |
Sziranyi T.,Computer and Automation Research Institute
Pattern Recognition Letters | Year: 2012
Deformable active contour (snake) models are efficient tools for object boundary detection. Existing alterations of the traditional gradient vector flow (GVF) model have reduced sensitivity to noise, parameters and initial location, but high curvatures and noisy, weakly contrasted boundaries cause difficulties for them. This paper introduces two Harris based parametric snake models, Harris based gradient vector flow (HGVF) and Harris based vector field convolution (HVFC), which use the curvature-sensitive Harris matrix to achieve a balanced, twin-functionality (corner and edge) feature map. To avoid initial location sensitivity, starting contour is defined as the convex hull of the most attractive points of the map. In the experimental part we compared our methods to the traditional external energy-inspired state-of-the-art GVF and VFC; the recently published parametric decoupled active contour (DAC) and the non-parametric Chan-Vese (ACWE) techniques. Results show that our methods outperform the classical approaches, when tested on images with high curvature, noisy boundaries. © 2012 Elsevier B.V. All rights reserved.
Elmaraghy W.,University of Windsor |
Elmaraghy H.,University of Windsor |
Tomiyama T.,Technical University of Delft |
Monostori L.,Computer and Automation Research Institute |
Monostori L.,Budapest University of Technology and Economics
CIRP Annals - Manufacturing Technology | Year: 2012
This paper reviews the breadth of complexity of the design process, products, manufacturing, and business. Manufacturing is facing unprecedented challenges due to increased variety, market volatility and distributed global manufacturing. A fundamental residue of globalization and market uncertainty is the increasing complexity of manufacturing, technological and economic systems. The nature and sources of complexity in these areas are reviewed and complexity modeling and management approaches are discussed. Enterprises that can mitigate the negative aspects of complexity while managing its positives should thrive on the continuous change and increasing complexity. To reap these benefits in the future, manufacturing companies need to not only adopt flexible technical solutions but must also effectively innovate and manage complex socio-technical systems. © 2012 CIRP.
Baranyi P.,Budapest University of Technology and Economics |
Baranyi P.,Computer and Automation Research Institute |
Csapo A.,Budapest University of Technology and Economics |
Csapo A.,Computer and Automation Research Institute
Acta Polytechnica Hungarica | Year: 2012
In this paper, we provide the finalized definition of Cognitive Infocommunications (CogInfoCom). Following the definition, we briefly describe the scope and goals of CogInfoCom, and discuss the common interests between CogInfoCom and the various research disciplines which contribute to this new field in a synergistic way.
Gyorgy A.,Hungarian Academy of Sciences |
Kocsis L.,Computer and Automation Research Institute
Journal of Artificial Intelligence Research | Year: 2011
Local search algorithms applied to optimization problems often suffer from getting trapped in a local optimum. The common solution for this deficiency is to restart the algorithm when no progress is observed. Alternatively, one can start multiple instances of a local search algorithm, and allocate computational resources (in particular, processing time) to the instances depending on their behavior. Hence, a multi-start strategy has to decide (dynamically) when to allocate additional resources to a particular instance and when to start new instances. In this paper we propose multi-start strategies motivated by works on multi-armed bandit problems and Lipschitz optimization with an unknown constant. The strategies continuously estimate the potential performance of each algorithm instance by supposing a convergence rate of the local search algorithm up to an unknown constant, and in every phase allocate resources to those instances that could converge to the optimum for a particular range of the constant. Asymptotic bounds are given on the performance of the strategies. In particular, we prove that at most a quadratic increase in the number of times the target function is evaluated is needed to achieve the performance of a local search algorithm started from the attraction region of the optimum. Experiments are provided using SPSA (Simultaneous Perturbation Stochastic Approximation) and kmeans as local search algorithms, and the results indicate that the proposed strategies work well in practice, and, in all cases studied, need only logarithmically more evaluations of the target function as opposed to the theoretically suggested quadratic increase. © 2011 AI Access Foundation. All rights reserved.
Kovacs A.,Computer and Automation Research Institute
ICAPS 2013 - Proceedings of the 23rd International Conference on Automated Planning and Scheduling | Year: 2013
This paper proposes a new model and algorithm for task sequencing in remote laser welding in the automotive industry. It is shown that task sequencing (in which order to weld the seams) is strongly related to path planning (how the welding robot should move), therefore the two problems must be solved together, in an integrated way. The problem is modeled as a direct product of a traveling salesman and a path planning problem, and a tabu search algorithm is proposed for solving it. Computational experiments show that the proposed method leads to a substantial reduction in the cycle time of the welding operation compared to an earlier approach. Copyright © 2013, Association for the Advancement of Artificial Intelligence. All rights reserved.
Hangos K.M.,Computer and Automation Research Institute
Entropy | Year: 2010
In this paper, the structural properties of chemical reaction systems obeying the mass action law are investigated and related to the physical and chemical properties of the system. An entropy-based Lyapunov function candidate serves as a tool for proving structural stability, the existence of which is guaranteed by the second law of thermodynamics. The commonly used engineering model reduction methods, the so-called quasi equilibrium and quasi steady state assumption based reductions, together with the variable lumping are formally defined as model transformations acting on the reaction graph. These model reduction transformations are analysed to find conditions when (a) the reduced model remains in the same reaction kinetic system class, (b) the reduced model retains the most important properties of the original one including structural stability. It is shown that both variable lumping and quasi equilibrium based reduction preserve both the reaction kinetic form and the structural stability of reaction kinetic models of closed systems with mass action law kinetics, but this is not always the case for the reduction based on quasi steady state assumption. © 2010 by the author.
Benedek C.,Computer and Automation Research Institute
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013
In this paper we introduce a probabilistic approach for extracting object ensembles from various digital images used by machine vision applications. The proposed framework extends conventional Marked Point Process models by allowing corresponding entities to form coherent object groups, by a Bayesian segmentation of the population. A global optimization process attempts to find the optimal configuration of entities and entity groups, considering the observed data, prior knowledge, and local interactions between the neighboring and semantically related objects. The proposed method is demonstrated in three different application areas: built in area analysis in remotely sensed images, traffic monitoring on airborne Lidar data and optical inspection of printed circuit boards. © 2013 Springer-Verlag.
Pataki M.,Computer and Automation Research Institute |
Marosi A.C.,Computer and Automation Research Institute
Journal of Grid Computing | Year: 2013
Translated or cross-lingual plagiarism is defined as the translation of someone else's work or words without marking it as such or without giving credit to the original author. The existence of cross-lingual plagiarism is not new, but only in recent years, due to the rapid development of the natural language processing, appeared the first algorithms which tackled the difficult task of detecting it. Most of these algorithms utilize machine translation to compare texts written in different languages. We propose a different method, which can effectively detect translations between language-pairs where machine translations still produce low quality results. Our new algorithm presented in this paper is based on information retrieval (IR) and a dictionary based similarity metric. The preprocessing of the candidate documents for the IR is computationally intensive, but easily parallelizable. We propose a desktop Grid solution for this task. As the application is time sensitive and the desktop Grid peers are unreliable, a resubmission mechanism is used which assures that all jobs of a batch finish within a reasonable time period without dramatically increasing the load on the whole system. © 2012 Springer Science+Business Media B.V.
Molnar J.,Eötvös Loránd University |
Chetverikov D.,Computer and Automation Research Institute |
Fazekas S.,Computer and Automation Research Institute
Computer Vision and Image Understanding | Year: 2010
We address the problem of variational optical flow for video processing applications that need fast operation and robustness to drastic variations in illumination. Recently, a solution  has been proposed based on the photometric invariants of the dichromatic reflection model . However, this solution is only applicable to colour videos with brightness variations. Greyscale videos, or colour videos with colour illumination changes cannot be adequately handled. We propose two illumination-robust variational methods based on cross-correlation that are applicable to colour and grey-level sequences and robust to brightness and colour illumination changes. First, we present a general implicit nonlinear scheme that assumes no particular analytical form of energy functional and can accommodate different image components and data metrics, including cross-correlation. We test the nonlinear scheme on standard synthetic data with artificial brightness and colour effects added and conclude that cross-correlation is robust to both kinds of illumination changes. Then we derive a fast linearised numerical scheme for cross-correlation based variational optical flow. We test the linearised algorithm on challenging data and compare it to a number of state-of-the-art variational flow methods. © 2010 Elsevier Inc. All rights reserved.
Kovacs A.,Computer and Automation Research Institute
International Journal of Production Economics | Year: 2011
This paper addresses the problem of storage assignment in a warehouse characterized by multi-command picking and served by milkrun logistics. In such a logistic system, vehicles circulate between the warehouse and the production facilities of the plant according to a pre-defined schedule, often with multiple cycles (routes) serving different departments. We assume that a request probability can be assigned to each item and each cycle, which leads to a special case of the correlated storage assignment problem. A MIP model is proposed for finding a class-based storage policy that minimizes the order cycle time, the average picking effort, or a linear combination of these two criteria. Computational experiments show that our approach can achieve an up to 3638% improvement in either criterion compared to the classical COI-based strategy. © 2010 Elsevier B.V. All rights reserved.