a borítólapra  Súgó epa Copyright 
Acta polytechnica HungaricaVolume 17., Issue No. 6. (2020.)

Tartalom

  • Gábor Gosztolya :

    Abstract: Automatically determining speaker emotions in human speech is a frequently studied task, where various techniques have been employed over the years. An efficient method is to represent the utterances by employing the Bag-of-Audio-Words technique, inspired by the Bag-of-Visual-Words approach from the area of image processing. In the past few years, however, Bag-of-Visual-Words has been replaced by the so-called Fisher vector representation, as it was shown to give a better classification performance. Despite this, in audio processing, Fisher vectors to date have only been rarely applied. In this study, we show that Fisher vectors are also a viable way of representing features in speech technology; more precisely, we use them in the task of emotion classification. Based on our results on two datasets, Fisher vectors can be effectively employed for this task: we measured 4% relative improvements in the UAR scores for both corpora, which rose to 9-16% when we combined this approach with the standard paralinguistic one.

    Keywords: audio processing; emotion detection; Fisher vector representation; Support Vector machines

  • Arsit Boonyaprapasorn ,
    Suwat Kuntanapreeda ,
    Teerawat Sangpet ,
    Parinya Sa Ngiamsunthorn ,
    Eakkachai Pengwang :

    Abstract: Pest management, based on biological control, has drawn attention from several research groups, due to the exclusion of chemical pesticides, which have debilitating outcomes, both on the environment and human health. Biological pest control policies have been determined using the model-based control approach. In this study, the tensor product model transformation (TPMT) was applied to model the nonlinear dynamic of the biological pest control system. Consequently, the feedback control law representing the biological pest control policy was synthesized based on LMI. Under the designed controller, the pest population was regulated based on the desired level. The simulation of the biological pest control system was presented to confirm the performance of the designed control law. It is evident, that the feedback control method based on TPMT can be employed appropriately, in this application.

    Keywords: tensor product model transformation method; biological control; pest management; feedback control

  • Imre Kiss :

    Abstract: This study presents several key aspects of Adamite hyper-eutectoid steel roll manufacturing. Using the multivariate statistical research used as modelling approach upon the industrial data (roll’s foundry and rolling sectors), the combined behavior of several chemical elements (graphitizing forming elements, Nickel |Ni| and Silicon |Si|), and carbure forming elements, Manganese |Mn| and Chrome |Cr|, under the presence of the Carbon |C|) upon the roll’s hardness are presented. In this sense, several results of a complex study on the Adamite hyper-eutectoid steel rolls are presented. For generating the multiple regression equations, determination of the specific correlation coefficients and drawing the graphical addenda, the software Matlab was used.

    Keywords: hyper-eutectoid steels; Adamite type rolls; graphitizing forming elements; carbide forming elements; hardness; multivariate regression analysis; correlation charts

  • Víctor C. da S. Campos ,
    Márcio F. Braga ,
    Luciano Frezzatto :

    Abstract: This work provides analytical upper bounds on the discretization error of uncertain linear systems. The Tensor Product Model Transformation is used to approximate the derived discretized system, with a reduced number of vertices. Digital state feedback controllers are then designed for the discretized system, for comparison to other available work in the current literature.

    Keywords: Tensor Product Model Transformation; Discretization; Uncertain Linear Systems

  • Judit Lukács :
    A Fuzzy Approach for In-Car Sound Quality Prediction75-94en [356.47 kB - PDF]EPA-02461-00101-0050

    Abstract: Numerous methods exist to characterize product quality. Nowadays, in the case of road vehicles, one of the most important issues is the acoustic comfort of the interior. However, the detection of the traffic environment is a further key question. In the case of minor vehicle collisions, the perceptibility is to analyze. Within the framework of the current study, the results of airborne noise measurements are presented. Experimental data were used to design predictive fuzzy models to estimate cabin noise level, which is in connection with the audibility of outer sourcing sounds. Two concepts of inference systems were investigated by examining accuracy, conformity and 0 residuals: Mamdani and Sugeno type ones. It was finally concluded that for estimating interior noise, Sugeno type fuzzy model is the better choice, as the accuracy and conformity are higher. In addition, the range of residuals is a magnitude lower: Mamdani type FIS provided -2.30 ~ 2.30 dB (-3.84 ~ 3.30%), Sugeno type one resulted -0.40 ~ 0.20 dB (-0.57 ~ 0.33%). Furthermore, the residuals follow a Gaussian distribution, in the case of the Sugeno predictive fuzzy model.

    Keywords: minor vehicle collisions; accompanying sound phenomenon; acoustic perception; airborne sound; vehicle interior noise; sound quality; pink noise; noise prediction; Mamdani-type FIS; Sugeno-type FIS

  • Marijo ©undrica ,
    Miroslav Petrinić :

    Abstract: This paper proposes a novel control system for synchronous machine rotor speed control. Based on deterministic observer and cascaded loops, a linear control system is obtained. Stability proof of the observer and linear analysis of the control system is given. Simulation studies of the machine starting, speed reversal and step loading are presented. A Comparison Analysis with the nonlinear control method, is also presented.

    Keywords: synchronous machine; observer; linear control; nonlinear control

  • Esteban Alfaro-Cortés ,
    José-Luis Alfaro-Navarro ,
    Matías Gámez ,
    Noelia García :
    Using Random Forest to Interpret Out-of-Control Signals115-130en [799.44 kB - PDF]EPA-02461-00101-0070

    Abstract: Statistical quality control procedures have become essential practices to ensure competitiveness in any manufacturing process. Since the quality of manufactured goods usually depends on several correlated characteristics, statistical multivariate techniques are needed to detect and analyze out-of-control situations. The difficulties in the interpretation of those out-of-control observations in multivariate control charts have motivated the development of different techniques in order to determine the variable or variables that have motivated the changes in the process and, in case of more than one variable as responsible of the change, to evaluate their contribution. Specifically, these techniques are mainly based in two alternatives, one that considers the T2 decomposition and other related to the application of classification techniques. The application of this latest techniques includes increasingly sophisticated methods, being the most usual alternative based on the application of Artificial Neural Networks. In this paper, we propose Random Forest as a powerful classification technique in statistical process control, considering a wide range of different situations in the function of the type of change and the magnitude of the correlation coefficient between variables. Moreover, the performance of Random Forest is analyzed in comparison with the results obtained from the application of Artificial Neural Networks to try to find out in which cases the superiority of Random Forest can be supported.

    Keywords: Hotelling T2; out-of-control; signals interpretation; Random Forest; Artificial Neural Networks

  • Dávid Csercsik :

    Abstract: In this article a multi-zonal integrated energy-reserve market model is proposed. Bidders in this model may submit their demand and supply bids on the one hand in the form of conventional hourly step bids and block bids, which are cleared and paid according to market clearing prices (MCPs). On the other hand, suppliers may submit so called flexible production bids, while both suppliers and consumers may submit fill-or-kill type packagepriced combined bids – these bids are accepted if their acceptance implies an improvement in the resulting total social welfare, which the market clearing algorithm aims to optimize. The model includes network constraints for the nominal case (if no reserves are activated) and also for perturbed cases when the allocated reserves are activated.

    Keywords: Integrated markets; Co-optimization; Market coupling; Market design

  • Tingting Wang ,
    Yanyun Bi ,
    Teng Hou ,
    Bo Liu ,
    Jianfeng Cui :

    Abstract: This paper presents a robust visual servoing controller based on an efficient TP model transformation method, while taking into account an uncertain image Jacobian matrix where, the camera intrinsic parameters, image features, and depth estimations are affected by unknown random uncertainties with known bounds. The convex vertex decomposition of image Jacobian matrix through uniform design greatly reduce the number of LMIs in the quasi-min-max model predictive control (MPC) scheme, in order to obtain the optimal control inputs of the constrained visual servoing system, while meeting the real-time requirements. Simulation and Experimental results demonstrate the effectiveness of the proposed method.

    Keywords: TP model transformation; Uniform design; Uncertain data; quasi-min-max MPC; LMIs

  • Miklós Póth ,
    ®eljen Trpovski ,
    Tatjana Lončar-Turukalo :

    Abstract: JPEG (Joint Photographic Experts Group) compression is the global standard for digital image compression introduced in 1992, and is still wide spread use. However, at low bitrates the JPEG process can introduce unwanted visual artifacts such as the blocking effects or edge ringing. This paper describes a method for modification and customizing of the JPEG compression. A nonlinear relationship between the quantization matrix, reflecting the compression ratio and the peak signal-to-noise ratio (PSNR), as an objective quality measure, was experimentally determined. The estimation of the quantization matrix and approximation of its mapping to a PSNR is accomplished relying on transformation of eleven test images using all quantization matrices. The linear approximation to this relation in the region of interest was proposed enabling fine tuning of the reconstructed image quality by either selection of the desired PSNR value or a decompressed image quality. In the decompression phase, post-processing is applied to reduce the blockish artifacts introduced by the compression process. The image block boundaries are first classified for an automatic identification of high blockiness, to constrain the application of a pre-processing algorithm and further loss of an image detail. Upon the reconstruction, the quality of the reconstructed image is measured using the PSNR and structural similarity index (SSIM). The effects of compression on the spectral properties are analyzed by comparison of the original and decompressed image spectra.

    Keywords: JPEG compression; digital image; discrete cosine transform; artifact reduction; block boundary classification; quantization; compression ratio

  • Aizhan Tlebaldinova ,
    Natalya Denissova ,
    Olga Baklanova ,
    Iurii Krak ,
    György Györök :

    Abstract: This paper presents technique for recognizing license plates structured characters of the Republic of Kazakhstan. This technique includes methods for converting the geometric-topological characteristics of license plates and the method for classifying alphanumeric characters by using cluster analysis. Developed modified algorithm for character recognition based on methods of contour analysis and template method with the addition of proposed transformations.

    Keywords: image processing; geometric-topological characteristics; contour analysis; character segmentation; character recognition

  • Faruk Selimović ,
    Predrag Stanimirović ,
    Muzafer Saračević ,
    Aybeyan Selimi ,
    Predrag Krtolica :

    Abstract: This paper presents the authentication method using the Delaunay triangulation incremental algorithm and the Catalan objects. The proposed method is a combination of computational geometry and cryptography. This method presents a new step towards encoding the triangle coordinates using the Catalan-key. We provided specific suggestions for the application of this method in the authentication for bank clients by the image encryption. Client authentication verification is performed by asking the client to enter the (x,y) coordinate values of randomly selected indices of an array. If the entered coordinates match the index values in the banking system array, then the transaction or other operation is approved. If the matching fails, it means that we have an unidentified person who has followed the whole process and wants to break into the banking system. There are many advantages arising from a scenario for the user authentication by the assigned Catalan object and the stack permutation method. Also, we provided concrete examples for the Delaunay encryption of image with an authentication scenario and experimental results for the proposed method.

    Keywords: Authentication; Cryptography; Delaunay triangulation; Catalan objects; Image encryption

  • Attila Geröcs ,
    Zoltan-Iosif Korka ,
    István Bíró ,
    Dorian Nedelcu :

    Abstract: The present paper is dedicated to the investigation of the kinematic and dynamic behavior of an inertial propulsion system, using rotating masses. The analytical results were compared with the results of a simulation using the SolidWorks Motion software. The following elements were considered: positional analysis, kinematic analysis of velocities and dynamic analysis.

    Keywords: dynamic analysis; inertial drive; kinematic; SolidWorks software

  • Zuzana Bilanová ,
    Ján Perháč ,
    Eva Chovancová ,
    Martin Chovanec :
    Logic Analysis of Natural Language Based on Predicate Linear Logic239-252en [784.59 kB - PDF]EPA-02461-00101-0140

    Abstract: This work discusses the formalization of sentence composition and the discovery of the semantic ambiguities of natural language. It also discusses the original connection between the logic area represented by predicate linear logic and ludics, as well as, the linguistic area represented by the Montague grammar. Montague grammar is a linguistic tool that allows analysing sentences in their extensional and intentional contexts. Predicate linear logic is a non-traditional logic of actions and resources where assumptions are consumed after the use of linear implication. Ludics uses proofs from predicate linear logic to analyse the strategies of actors in dialogues. The contribution of this work is to practically demonstrate this approach by translating a natural-language sentence into a predicate linear formula and describe it in time-spatial calculus.

    Keywords: ludics theory; Montague grammar; predicate linear logic