Andras Horvath

57198324582

Publications - 19

Applicability of a computer retinal model for scale-dependent investigation of legibility problems

Gábor Dömötör Andras Horvath

Publication Name: Heliyon

Publication Date: 2025-02-28

Volume: 11

Issue: 4

Page Range: Unknown

Description:

Selecting an appropriate typeface is crucial in typography, web design, and other applications where text readability is essential. A key concept of this topic is legibility, the quality that shows how easy it is to recognize the letters of a particular font set. Previous works have measured legibility by human experiments, which has several limitations; for example, the methodology and circumstances were not entirely uniform, and the results may be distorted by the fatigue of the test subjects. This paper presents a new method using self-developed software to substitute human measurements in legibility testing. The software simulates the human retina's distortion effects (direction-dependent acuity) on test images of the assessed font's characters. Then, its output is analyzed using optical character recognition software. By integrating these techniques, we model the optical, biological, and cognitive steps of human character recognition as well. Although the simulation is imperfect, the software can perform significantly more measurements than human experiments with higher uniformity and give reproducible legibility information about significantly more fonts in various circumstances. In addition to the two scaling methods used in the literature (x-height, font-height), the tests are also performed with two other self-developed scaling methods, which provide a fairer comparison in the case of non-standard character types. This paper contains the legibility measurement results for 22 fonts under various simulated scenarios. The derived font ranking aligns closely with findings from prior human-based studies, demonstrating the robustness and reliability of the proposed method. Moreover, this approach provides valuable insights into font legibility across a broader spectrum of use cases, highlighting its potential for practical applications in typography and design.

Open Access: Yes

DOI: 10.1016/j.heliyon.2025.e42484

Adaptive Vehicle Trajectory Clustering Based on Computer Vision

Andras Horvath Áron Dávid Agg Bence Gábor Péter

Publication Name: Lecture Notes in Networks and Systems

Publication Date: 2025-01-01

Volume: 1258 LNNS

Issue: Unknown

Page Range: 344-360

Description:

Intelligent Transportation Systems is a rapidly evolving and extensively researched field. The development of tools and methods necessitated the appearance of a new concept, Cognitive Mobility, covers greater integration of transport related areas. Our research goes beyond simple image processing and statistical analysis and includes cognitive elements of mobility, for example, gives methods for adaptive analysis of typical vehicle behavior in junctions and helps in decision making for better utilization of infrastructure. In this work, we present an adaptive approach to interpreting traffic scenes. This approach uses the YOLOv7 object detector and DeepSORT tracking algorithm to generate trajectories of moving vehicles in video recordings captured from a stationary viewpoint. The paper examines and compares various techniques for clustering the vehicle trajectories, such as using clustering algorithms to group vehicles based on their entry and exit points within the video scene. Furthermore, we report the results of clustering parameter sensitivity tests performed on the video dataset we created. This control over the granularity of the clustering enables generating street-level or lane-level traffic data from the same recordings. Finally, we demonstrate some practical applicability of the method through illustrative examples, showcasing how the generated traffic data can be used to monitor and analyze traffic patterns, identify congestion points, and inform transportation planning and decision-making.

Open Access: Yes

DOI: 10.1007/978-3-031-81799-1_32

Exploring disparities in the generation of food waste from a spatial and sustainability perspective

A. Zseni Andras Horvath Gergely Zoltán Macher Dóra Sipos Judit Pécsinger

Publication Name: Bio Web of Conferences

Publication Date: 2024-08-23

Volume: 125

Issue: Unknown

Page Range: Unknown

Description:

In the near future, food production will encounter numerous challenges. The world's population, per capita intake of calories, protein, and cereals is growing, and expected to continue increasing in the future. According to the World Resources Institute, feeding 9-10 billion people by 2050 will require a 70% increase in food calorie compared to 2006 levels. The expansion of crop and livestock production, the increasing use of fertilisers and pesticides at global level, and the growing adoption of precision farming methods may not be appropriate in the light of problems such as soil degradation, the impact of climate change on yields, the depletion of fish stocks and the reduction of arable land per capita in certain areas. Mitigation of food waste can be a possible way to sustainable food supply but it alone cannot solve the problem. The paper focuses on examining disparities in the generation of food waste from a spatial and sustainability perspective within EU-countries. Data related to food waste and various environmental, social and economic indicators are collected for EU countries. Correlation between these parameters is investigated to find possible connections. Cluster analysis is applied to food waste data to find the groups of countries with similar characteristics. The distribution of environmental, economical, and social parameters is calculated for these groups, and statistical methods are applied to investigate their differences.

Open Access: Yes

DOI: 10.1051/bioconf/202412502002

Response to Artificial intelligence-based colorectal polyp histology prediction using narrow-band image-magnifying colonoscopy: a stepping stone for clinical practice

István Rácz Andras Horvath Zoltán Horváth

Publication Name: Clinical Endoscopy

Publication Date: 2022-09-01

Volume: 55

Issue: 5

Page Range: 701-702

Description:

No description provided

Open Access: Yes

DOI: 10.5946/ce.2022.123.1

MEASURING AGE-DEPENDENCE OF COLOUR AFTERIMAGE PERCEPTION

Lorinc Garai Andras Horvath

Publication Name: Light and Engineering

Publication Date: 2022-01-01

Volume: 30

Issue: 2

Page Range: 70-81

Description:

Afterimages are common and frequent perceptual phenomena of everyday life. A typical appearance is the negative “ghost” image of a bright light source when we turn away from it. In the case of significant colour contrast, the afterimage can be coloured. The perceived false image’s strength decreases gradually and completely disappears in a (10–100) s timescale. The underlying processes have multiple components: a quick adaptation on the retinal level, and a slower adaptation on the neural level. Several studies discuss these mechanisms, but there are still important questions to be answered. In our research, we apply the top-level, black-box style approach: instead of focusing on the inner details, we ask human test subjects to test and measure the duration and “strength score” of the same light-transitions. Our goal is to find the main features that affect the duration and subjective strength of the colour afterimages. Specifically, we examine whether the age and gender of the test subjects or the colourimetry parameters affect these parameters. Two set of experiments were performed: colour-colour transitions with 41 and colour-grey transitions with 16 test subjects between 19 and 62. We found that gender has no measurable influence, but age makes a difference in high significance. Both experiment types confirmed that over 40 years the average duration of colour afterimages decreases.

Open Access: Yes

DOI: 10.33383/2021-061

Artificial Intelligence-Based Colorectal Polyp Histology Prediction by Using Narrow-Band Image-Magnifying Colonoscopy

István Rácz Henriett Regoczi Noemi Kranitz Andras Horvath Gyöngyi Kiss Zoltán Horváth

Publication Name: Clinical Endoscopy

Publication Date: 2022-01-01

Volume: 55

Issue: 1

Page Range: 113-121

Description:

Background/Aims: We have been developing artificial intelligence based polyp histology prediction (AIPHP) method to classify Narrow Band Imaging (NBI) magnifying colonoscopy images to predict the hyperplastic or neoplastic histology of polyps. Our aim was to analyze the accuracy of AIPHP and narrow-band imaging international colorectal endoscopic (NICE) classification based histology predictions and also to compare the results of the two methods. Methods: We studied 373 colorectal polyp samples taken by polypectomy from 279 patients. The documented NBI still images were analyzed by the AIPHP method and by the NICE classification parallel. The AIPHP software was created by machine learning method. The software measures five geometrical and color features on the endoscopic image. Results: The accuracy of AIPHP was 86.6% (323/373) in total of polyps. We compared the AIPHP accuracy results for diminutive and non-diminutive polyps (82.1% vs. 92.2%; p=0.0032). The accuracy of the hyperplastic histology prediction was significantly better by NICE compared to AIPHP method both in the diminutive polyps (n=207) (95.2% vs. 82.1%) (p<0.001) and also in all evaluated polyps (n=373) (97.1% vs. 86.6%) (p<0.001) Conclusions: Our artificial intelligence based polyp histology prediction software could predict histology with high accuracy only in the large size polyp subgroup.

Open Access: Yes

DOI: 10.5946/ce.2021.149

Colour afterimages: Computing and measuring virtual colour perception provided by two display types

Lorinc Garai Andras Horvath

Publication Name: 11th IEEE International Conference on Cognitive Infocommunications Coginfocom 2020 Proceedings

Publication Date: 2020-09-23

Volume: Unknown

Issue: Unknown

Page Range: 109-114

Description:

Afterimages are common and frequent perceptual phenomena of everyday life. When looking into a bright light source compared to environment and suddenly turning the eyes away from it, the illusionary image of the light source remains in perception, for a while. A subordinate to afterimages defined in our research is virtual colour perception that, according to our interpretation is an unusual and intense temporary colour perception provoked by quick colour change in the incident light. Modelling virtual colour perception can be useful in the field of visual design, for example in dynamic advertising. A computational kinetic model established for simulating and quantifying virtual colour perception was developed in our previous work and was validated with preliminary tests involving 20 observers. Validation test results were contestable due to the high variance (about 50 RSD%) of duration and intensity of afterimage results at each incident light colour change. Thus, in our current work, validation on one-observer tests was performed to minimize the variance of results. Further, these validation tests were run on two liquid crystal display (LCD) types: a nowadays frequently present In-Plane Switching (IPS) display providing vivid colours, and an older LCD-TN (twisted nematic) display.

Open Access: Yes

DOI: 10.1109/CogInfoCom50765.2020.9237835

Measuring colour afterimage perception

Lorinc Garai Andras Horvath

Publication Name: 11th IEEE International Conference on Cognitive Infocommunications Coginfocom 2020 Proceedings

Publication Date: 2020-09-23

Volume: Unknown

Issue: Unknown

Page Range: 191-196

Description:

Afterimages are common and frequent perceptual phenomena of everyday life. When looking into a high-intensity light source and suddenly turning away from it, a temporary 'ghost' of the light source remains in perception, for a while. A highly saturated, coloured incident light has a different effect on the three types of cone cells. Therefore, the fast change in the colour of incident light triggers different intensity afterimage on the different types of cones, and the observer perceives a colour afterimage disappearing in a between 10- and 100-second timescale. By choosing incident light colours, colour afterimages can be applied in visual design, like in modern, dynamic advertising. Our previous work describes simulation of colour afterimages triggered by given colour combinations. For the validation of simulation, preliminary tests were performed with 20 subjects. Test results suggested the possibility of age dependence of intensity and duration of colour afterimages. In this work, we present the experimental results of 28 observers with different ages and gender. The disappearing time and perceived initial 'intensity' of the colour afterimages were measured in nine fast colour transition cases.

Open Access: Yes

DOI: 10.1109/CogInfoCom50765.2020.9237851

A new assessment methodology for fair-faced concrete surfaces based on digital image processing

Andras Horvath Kitti Karolyfi Ferenc Papp

Publication Name: Proceedings of the 2020 Session of the 13th Fib International Phd Symposium in Civil Engineering

Publication Date: 2020-01-01

Volume: Unknown

Issue: Unknown

Page Range: 288-295

Description:

Fair-faced concrete elements have become even more popular in architecture because of their pure aesthetics and favourable material properties. The europian standards and guidelines are usually specifying four classes regarding to the surface quality. However, the strandard assessment methodologies are based only on manual methods and they are very subjective in some cases. In order to increase the efficiency and accuary of the evaluation a new methodology is proposed using digital image processing. In this paper the basis of the evaluation method made by using the Python 3.6 software is presented by focusing on the surface void ratio which is one of the most common surface defects of fair-faced concrete structures.

Open Access: Yes

DOI: DOI not available

Modelling of human color perception depending on quick color shifts on screen

Lorinc Garai Andras Horvath

Publication Name: 9th IEEE International Conference on Cognitive Infocommunications Coginfocom 2018 Proceedings

Publication Date: 2018-07-02

Volume: Unknown

Issue: Unknown

Page Range: 223-226

Description:

The eye is considered as the most important human sensory organ. Its important part is the retina, which converts light to neural impulses by several millions of sensor cells: rods and cones. To understand this process, modelling by computer simulation is required. The well-known phenomenon of afterimages is examined in this paper. When a fast color change on screen is performed, a color after image appears temporarily. This afterimage may result in a more intensive color perception. A model was established to characterize the color perception intensity and duration depending on color changes.

Open Access: Yes

DOI: 10.1109/CogInfoCom.2018.8639909

Computer model for font legibility measurement

Gábor Dömötör Andras Horvath

Publication Name: Intelligent Decision Technologies

Publication Date: 2017-01-01

Volume: 11

Issue: 3

Page Range: 335-344

Description:

Selecting a good typeface (or font) has a high importance in typography, web-design, ergonomics and similar fields. A key concept of this topic is the legibility of the fonts, which measures how easy it is to recognise the difference between the letters in a specific typeface. The typical investigations to measure this feature are based on human experiments. In this paper, a new approach and a simulation software are proposed to measure font legibility. A program, that simulates the distortion effects of the human retina is applied to test images of characters and optical character recognition software is used to investigate its output. This way the optical, biological, and cognitive steps of human character recognition is modelled. Although, the simulation cannot be perfect, the computer program can investigate much more fonts, characters with a lot of settings (e.g. for different visual acuity) than human experiments and gives reproducible results that are independent of fatigue, misunderstanding and other human limits.

Open Access: Yes

DOI: 10.3233/IDT-170299

Spectral reconstruction from tristimulus values with the use of principal component analysis and genetic optimization

Zsolt Sávoli Andras Horvath Balázs Kránicz

Publication Name: Light and Engineering

Publication Date: 2017-01-01

Volume: 25

Issue: 1

Page Range: 96-105

Description:

Reconstruction of spectrum or spectral reflectance using only tristimulus values is impossible because of the infinite number of metamers. However, with additional information about the type of light source or surface, approximate reconstruction is possible. Several studies were written on this topic using Principal Component Analysis (PCA): using a sample set of spectra (or spectral reflectance) the search space for appropriate metamer can be narrowed. In this paper a new PCA-based optimization method is presented, which is more accurate than that of published before and gives non-negative, smooth results. Using precalculated lookup tables, the time and memory consumption will be very small. This method can produce a good reconstructed, realistic spectrum of an illuminant or spectral reflectance function for a surface based on tristimulus values of a digital image.

Open Access: Yes

DOI: DOI not available

Computational simulation of mesopic vision based on camera recordings

Gábor Dömötör Andras Horvath

Publication Name: Light and Engineering

Publication Date: 2014-01-01

Volume: 22

Issue: 1

Page Range: 61-67

Description:

Understanding mesopic vision is one of the hottest topics in lighting engineering [1]. The high importance of the field comes from the wide variety of the applications such as driving and other outdoor activities are performed in mesopic conditions at night. Therefore, it is extremely important to calculate the visibility of different objects in such conditions. In this paper, we present the algorithm and the first results of software, which is able to simulate the behaviour of human retina in scotopic, mesopic and photopic ranges, calculates the direction-dependent acuity. Our aim was to produce a practical tool to examine real-time situations. Therefore, classical camera recordings are used as input, which limits the accuracy but extends the range of applications.

Open Access: Yes

DOI: DOI not available

Comparative analysis of parallel gene transfer operators in the bacterial evolutionary algorithm

F. Hatwágner Andras Horvath

Publication Name: Acta Polytechnica Hungarica

Publication Date: 2012-09-18

Volume: 9

Issue: 4

Page Range: 65-84

Description:

The Bacterial Evolutionary Algorithm (BEA) is an evolutionary method, originally meant to optimize the parameters of fuzzy systems. The authors have already proposed three modified versions of the original algorithm in a previous paper to make it usable in engineering applications with time-consuming object functions as well. Section 1 summarizes the earlier results. It presents the operators of the original BEA and the suggested parallel version. In Section 2, the optimal parameter settings and the analytical estimation of wall clock time in parallel computations are investigated. In Section 3, the paper deals with genetic diversity in different BEA versions. The effect of the modified gene transfer operators on genetic diversity is measured. The conclusion is that the proposed methods have quite good efficiency in all cases, and we can reach the ideal case if we have full control over the parameters.

Open Access: Yes

DOI: DOI not available

Searching for a nonlinear ODE model of vehicle crash with genetic optimization

F. Hatwágner István Harmati Andras Horvath

Publication Name: Saci 2012 7th IEEE International Symposium on Applied Computational Intelligence and Informatics Proceedings

Publication Date: 2012-01-01

Volume: Unknown

Issue: Unknown

Page Range: 131-136

Description:

Vehicle crash is a very complex process, which can be modelled in details using the finite element method (FEM), but a simple, quasi-heuristic model with a limited number of parameters is often more beneficial. In this paper we propose a relatively simple dynamic model for deformation and force during a frontal collision process, which has very similar behavior to the experimental data. A genetic-type optimization of model parameters is executed on three car crash experimental data sets. ©2012 IEEE.

Open Access: Yes

DOI: 10.1109/saci.2012.6249990

Error handling techniques of genetic algorithms in parallel computing environment

F. Hatwágner Andras Horvath

Publication Name: Pollack Periodica

Publication Date: 2008-08-01

Volume: 3

Issue: 2

Page Range: 3-14

Description:

It is easy to create parallel genetic algorithm software with master-slave type paralelization on a cluster of workstations. In a real situation the probability of errors in communication or in some of the slave processes during a long calculation is significant. In this paper we deal with different error handling strategies in master-slave type paralelization of standard GA algorithms and show results of test calculations. Our simulations are close to real applications in the sense that we examine the best achieved objective function value at a fixed wall clock time with different error handling strategies depending on the probability of errors and number of processors. Using these results we make suggestions on the selection of a good error handling method in different optimization problems. © 2008 Akadémiai Kiadó.

Open Access: Yes

DOI: 10.1556/Pollack.3.2008.2.1

The effect of computer network errors on genetic algorithms

F. Hatwágner Andras Horvath

Publication Name: Pollack Periodica

Publication Date: 2007-08-01

Volume: 2

Issue: 2

Page Range: 3-12

Description:

Genetic algorithms are widely used in engineering, to solve nonlinear, multi-target optimization problems with multiple variables (e.g. optimization of geometry of flow domains, parameters of control systems). The parallelization of software using genetic algorithms is very important because in a typical practical problem they need huge computational power. Fortunately it is easy to implement a master-slave style parallelization. Our goal was to investigate the effect of random errors that can occur in a cluster of workstations on the efficiency of the genetic algorithm. © 2007 Akadèmiai Kiadó.

Open Access: Yes

DOI: 10.1556/Pollack.2.2007.2.1

A new assessment methodology for fair-faced concrete surfaces based on digital image processing

Andras Horvath Kitti Karolyfi Ferenc Papp

Publication Name: Fib Symposium

Publication Date: 2020-01-01

Volume: Unknown

Issue: Unknown

Page Range: 288-295

Description:

Fair-faced concrete elements have become even more popular in architecture because of their pure aesthetics and favourable material properties. The europian standards and guidelines are usually specifying four classes regarding to the surface quality. However, the strandard assessment methodologies are based only on manual methods and they are very subjective in some cases. In order to increase the efficiency and accuary of the evaluation a new methodology is proposed using digital image processing. In this paper the basis of the evaluation method made by using the Python 3.6 software is presented by focusing on the surface void ratio which is one of the most common surface defects of fair-faced concrete structures.

Open Access: Yes

DOI: DOI not available

Using Multivariate Statistical Analysis for Examining the Relationship between Food Waste Generation and Socio-economic Factors

A. Zseni Andras Horvath Gergely Zoltán Macher Dóra Sipos Judit Pécsinger

Publication Name: Journal of Sustainable Development of Energy Water and Environment Systems

Publication Date: 2025-09-01

Volume: 13

Issue: 3

Page Range: 1-16

Description:

Food waste contributes to social inequalities and sustainability issues by worsening resource overuse and environmental harm. The United Nations Sustainable Development Goal 17 highlights the importance of reducing food waste to address hunger and promote a sustainable, economically viable global food system. This paper examines the geographic differences in food waste levels among European Union member nations and analyses the associations between food waste and diverse environmental, geographic, social and economic indicators, including Sustainable Development Goals and other sustainability metrics. Using dimensionality reduction methods, nontrivial multivariate connections between food waste and these parameters were identified, allowing for the characterisation of countries based on a few significant factors. Principal Component Analysis (PCA), applied to food waste data across European Union countries, uncovered three distinct groups: (1) those with elevated food waste in primary production, manufacturing and distribution stages; (2) those with lower waste in these domains yet elevated waste in restaurants and households; and (3) those with all of their food waste components smaller than or equal to the average. The multivariate linear correlation between the PCA factors and socio-economic parameters is nonsignificant, but a few (nonlinear) regularities could be identified: five of the six countries of the first group above are characterised by the population settled mainly on flatland and an above-average supply of meat or fish. Another pattern observed is that former Eastern Bloc countries belong to the third group. The research findings offer valuable insights that can inform the efforts of environmental experts, professionals and policymakers working in the circular economy and waste management domains. This knowledge can facilitate the development of more effective strategies aimed at mitigating food waste and promoting sustainability.

Open Access: Yes

DOI: 10.13044/j.sdewes.d13.0579