Search Everything

Tip: Search using "First Name + Last Name", e.g.
János Kiss instead of Kiss János.

Publications - 6374

Rail Defect Classification with Deep Learning Method

Publication Name: Acta Polytechnica Hungarica

Publication Date: 2022-01-01

Volume: 19

Issue: 6

Page Range: 225-241

Description:

The good condition of railway rails is crucial to ensuring the safe operation of the railway network. At present, the rail flaw detectors are widely used in rail flaw detection, they are typically based on the principle of ultrasonic detection. However, the rail detection results analysis process involves huge manual work and the associated labor costs, with low levels of efficiency. In order to improve the efficiency, accuracy of results analysis and also reduce the labor costs, it is necessary to employ classification of ultrasonic flaw detection B-scan image, based on an artificial intelligence algorithm. Inspired by transformer models, with excellent performance in the field of natural language processing (NLP), some deep learning models differ from traditional convolutional neural networks (CNN), gradually emerge in the field of computer image processing. In order to explore the practicality of this model in the field of computer image processing (vision), in the paper, the Vision Transformer (ViT) is employed to train with rail defect B-scan images data and produce a rail defect classification. The model accuracy is more than 90% with the highest accuracy reaching 98.92%.

Open Access: Yes

DOI: DOI not available

Past, present, and future of block-chain in finance

Publication Name: Journal of Business Research

Publication Date: 2024-04-01

Volume: 177

Issue: Unknown

Page Range: Unknown

Description:

Diverse businesses are investigating the possibility of redefining their current operational systems in light of the latest blockchain, initially developed for Bitcoin traBitcoinns. This research examines the existing literature on blockchain and its application in the finance sector. This paper provides a systematic literature review of the uses of blockchain in the finance sector. To conduct the review, we performed a boolean search on the Scopus database and obtained 149 records, which we then analyzed bibliometrically using the bibliometrix package in R. The categorization of the existing literature into themes resulted in identifying the following six significant research themes: financial inclusion, sustainable finance, blockchain technology, cryptocurrencies, and artificial intelligence. Following the inductive analysis, we propose a conceptual framework that includes components such as the digital financial revolution, innovation, entrepreneurship, the financial market, sustainable business development, and financial innovation and sustainability. These findings are utilized to suggest future lines of inquiry for this area of study, including the necessity of methodological development and theoretical foundation.

Open Access: Yes

DOI: 10.1016/j.jbusres.2024.114640

Effect of ethanol solvents on total phenolic content and antioxidant properties of seed extracts of fenugreek (Trigonella foenum-graecum l.) varieties and determination of phenolic composition by hplc-esi-ms

Publication Name: Diversity

Publication Date: 2022-01-01

Volume: 14

Issue: 1

Page Range: Unknown

Description:

Fenugreek (Trigonella foenum-graecum L.) is one of the oldest cultivated plants grown for its leaves and seeds that are used for both culinary and medicinal purpose. This study aims to evaluate the effect of ethanol concentration (30, 50, 70 and 96% (v/v) of ethanol in water) as a solvent for the extraction of total phenolic content (TPC) and antioxidant properties (antiradical activity (ARA), transition metal reducing power (TMRP), iron chelating ability (ICA)) of seed extracts of spring variety Ovari 4 (FSV) and winter variety PSZ.G.SZ (FWV) fenugreek, and separate and identify the major phenolics of the extracts by HPLC-ESI-MS. The results indicated that 70% ethanol solution resulted in the maximum amount of TPC for both FSV and FWV seeds. The TPC decreased in the treatments in the following order: 70% ethanol > 96% ethanol > 50% ethanol > 30% ethanol, whereas extraction yield changed in a different manner: 30% ethanol > 50% ethanol > 70% ethanol > 96% ethanol. The extracts from seeds of both fenugreek varieties obtained with 70% and 96% ethanol showed equal high RSA while superior TMRP and ICA were observed in 70% ethanol extracts. The TMRP and ICA were strongly correlated with TPC for both varieties. The correlation between RSA and TPC was high, but not significant. Thus, the obtained data indicate the 70% ethanol solvent suitability for efficient extraction of phenolic compounds from seeds of the FWV and FSV. According to an HPLC-ESI-MS analysis, the polyphenolic profiles of fenugreek are presumably formed by flavone C-glycosides with apigenin or luteolin as aglycone linked with different glycones. High antioxidant activity of FWV seeds can be an adaptation to cold stress of the winter variety aimed at strengthening the antioxidant defense of the germinating seeds.

Open Access: Yes

DOI: 10.3390/d14010007

Increasing relevance of non-human genetics in Hungarian forensic practice

Publication Name: Forensic Science International Genetics Supplement Series

Publication Date: 2015-12-01

Volume: 5

Issue: Unknown

Page Range: e250-e252

Description:

In step with the evolution of molecular genetics, the application of non-human biological traces is an intensively researched field in forensic sciences [1]. Recently it has been widely acknowledged that this extended possibility – with associated abilities and pitfalls – for forensic DNA examination can provide innovative results in criminal and civil casework as well as in connection with associative evidence. Depending on the specific species or types of traces, the final results can have very different criminalistics value (e.g., plants, dogs or microbiome of soil traces), but it is unquestionable that non-human DNA has an increasing role in forensic biological evidence [2–4]. However, in contrast to the investigation of human DNA-traces, there currently exist no restrictions concerning DNA loci and privacy of genetic information, and a wide spectrum of genetic information and technologies may be applied. The field is developing with animal rights, international (e.g. CITES) conventions, national regulations and professional recommendations [5–7]. The Hungarian forensic practice on animal DNA examination has been developing since 2000 [8]. The range of cases is sufficiently broad, and includes fatal accidents and animal attacks (case #1) as well as illegal trade cases (case #2). In case of degraded animal hair samples we used two sets of Canine STR markers with shortened amplicons, and species specific primers (Ursus sp.) for mitochondrial DNA analysis of mixed samples (case #2), respectively. The results supported the evaluation of individualization (case #1), confirming the species identification (case #2).

Open Access: Yes

DOI: 10.1016/j.fsigss.2015.09.100

Empirical identification of the creative cognitive unconscious processes in the collective individuation concerning the World-Clock models: Part II. Pauli's regiomontanus dream and its historical and spiritual background

Publication Name: 6th IEEE Conference on Cognitive Infocommunications Coginfocom 2015 Proceedings

Publication Date: 2016-01-25

Volume: Unknown

Issue: Unknown

Page Range: 607-626

Description:

This part of the paper first deals with some methodological problems of the possible identification of special patterns in the hypothetical evolution of the broadening reflective consciousness ("collective individuation") related to the spiritual history of the 11th century. Furthermore, we intend to present some hypothetical important episodes of the process of the development of the consciousness, the collective individuation with partial interpretation of Pauli's Regiomontanus dream and subsequent famous dream series of 1954-1956, by their interpretation as a long run cognitive process.

Open Access: Yes

DOI: 10.1109/CogInfoCom.2015.7390661

Empirical identification of the creative cognitive unconscious processes in the collective individuation concerning the World-Clock models: Part III. the three hidden pillars of the world and the fourth one

Publication Name: 6th IEEE Conference on Cognitive Infocommunications Coginfocom 2015 Proceedings

Publication Date: 2016-01-25

Volume: Unknown

Issue: Unknown

Page Range: 627-646

Description:

This part of the paper discusses after the broad areas (suiting the three background languages which were recommended by Pauli) of the 'mythological world', the 'physical cosmos' and the depth psychology, which gained a really worthy representation in the three great minds (Pauli, Jung and Kerényi), and the fourth one, bounded to the city of Zurich. The fourth key area is linked to the models of cosmic language mysticism and to the corresponding linguistic aesthetic background language. As a representative, ensuring the completeness of the quaternion, the Zurich linked James Joyce, who lived and deceased in Zurich, is seen, in a way in the creative unconscious evolutionary processes, as if he was in inseparable close unconscious contact with the other three persons.

Open Access: Yes

DOI: 10.1109/CogInfoCom.2015.7390662

A Systematic Review of Urban Mobility and Preliminary Research of Transportation Trends in West Hungary †

Publication Name: Engineering Proceedings

Publication Date: 2024-01-01

Volume: 79

Issue: 1

Page Range: Unknown

Description:

Based on the available statistical data, most employees travel by car, which results in traffic jams, hurts the environment, and increases household expenditures. As a result of this phenomenon, our investigation aimed to present the global trends in urban mobility. It is also considered preliminary research of Hungarian transportation habits and trends by summarizing the standard fees of taxi service, parking, and maintaining a driving license. As our research methodology, we chose a systematic literature search and a summary of the available secondary data to present the main global trends characterizing urban mobility and the standards typical of Hungarian transport. The results enable the expansion of the research and provide an opportunity to learn about consumers’ attitudes at the national level.

Open Access: Yes

DOI: 10.3390/engproc2024079007

Quasi optimization of fuzzy neural networks

Publication Name: 10th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics Cinti 2009

Publication Date: 2009-12-01

Volume: Unknown

Issue: Unknown

Page Range: 303-314

Description:

The fuzzy flip-flop based multilayer perceptron, named Fuzzy Neural Network, FNN is proposed for function approximation. In recent years much effort has been made for the development of a special kind of bacterial memetic algorithm for optimization and training of the fuzzy neural network parameters. In this approach the FNN parameters have been encoded in a chromosome and participate in the bacterial mutation cycle. The quasi optimized FNN's performance based on various fuzzy flip-flop types has been examined with a series of multidimensional input functions.

Open Access: Yes

DOI: DOI not available

STRESS ANALYSIS OF PRESSING JOINT OF TWO CONICAL PIPES

Publication Name: Strojnicky Casopis Journal of Mechanical Engineering

Publication Date: 2023-05-01

Volume: 73

Issue: 1

Page Range: 117-124

Description:

This paper presents a stress analysis of a pressed joint of two conical pipes with the aim to find out whether this kind of connection is suitable for telecommunication masts. The analysis can be considered a complex spatial problem of a thin-walled shell, where the local bending stress influence might be dominant. In this paper a solution using engineering approach and numerical analysis with finite element method (FEM) will be compared.

Open Access: Yes

DOI: 10.2478/scjme-2023-0009