Ladislav Choma

Dr. Ladislav Choma

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 June 2025 in Acta Avionica Journal
Verified
Journal Article
Open access
PDF file
LCLadislav Choma
Ladislav Choma
MKMartin Kelemen
Martin Kelemen
MAMatej Antoško
Matej Antoško
KOKristina Ozdincova
Kristina Ozdincova
JSJozef Sabo
Jozef Sabo
The prediction of hazardous weather phenomena is a critical component in ensuring the safety and efficiency of air transport operations. This paper focuses on evaluating the potential of artificial intelligence (AI) in forecasting fog—one of the most significant weather conditions affecting airport visibility. The methodological framework combines empirical methods (observation, measurement, experimentation) with theoretical approaches (analysis, synthesis, modeling). Emphasis is placed on the application of machine learning and deep learning techniques for processing meteorological data collected from Sliac military airport. The study compares conventional numerical weather prediction models (e.g., WRF) with AI-based approaches such as Support Vector Machines (SVM), Long Short-Term Memory (LSTM) neural networks, and ensemble models. Results indicate that AI models achieve higher accuracy in short-term fog prediction while reducing computational requirements. Experiments demonstrated success rates of up to 90% using ensemble techniques. The findings confirm that AI represents a promising tool for developing modern predictive meteorological systems in aviation. Challenges identified include limited data availability, the need for high-quality datasets, and the complexity of model interpretation. Future work should include expanding the data scope to multiple airports and incorporating satellite and radar data. The proposed approach offers a strong foundation for the advancement of intelligent, automated decision-support systems in both civil and military aviation meteorology.

29 July 2025 in Atmosphere
Verified
Journal Article
Open access
PDF file
LCLadislav Choma
Ladislav Choma
MAMatej Antoško
Matej Antoško
Peter KorbaPeter Korba
Peter Korba
This study focuses on surface wind analysis at the small regional airport in Svidnik, used primarily for pilot training under daytime VFR conditions. Due to the complex local terrain and lack of prior meteorological data, an automatic weather station was installed, collecting over 208,000 wind measurements over a two-year period at ten-minute intervals. The dataset was processed using hierarchical filtering and statistical selection, and visualized via wind rose diagrams. The results confirmed a dominant southeastern wind component, supporting the current runway orientation (01/19). However, a less frequent easterly wind direction was identified as a safety concern, causing turbulence near the runway due to terrain and vegetation. This is particularly critical for trainee pilots during final approach and landing. Statistical analysis showed that easterly winds, though less common, appear year-round with a peak in summer. Pearson correlation and linear regression confirmed a significant relationship between the number of easterly wind days and their measurement frequency. Daytime winds were stronger than nighttime, justifying the focus on daylight data. The study provides practical recommendations for training flight safety and highlights the value of localized wind monitoring at small airports. The presented methodology offers a framework for improving operational awareness and reducing risk in complex environments.

24 August 2026 in Meteorology and Atmospheric Physics
Verified
Journal Article
Open access
PDF file
HBHenrich Balaj
Henrich Balaj
LILukáš Ivica
Lukáš Ivica
LCLadislav Choma
Ladislav Choma
Miroslav KelemenMiroslav Kelemen
Miroslav Kelemen
JBJuraj Bartok
Juraj Bartok
IOIrina Malkin Ondík
Irina Malkin Ondík
The detection of dangerous convective cloud types, specifically cumulonimbus and towering cumulus, is crucial for aviation safety and meteorological forecasting. Traditionally reliant on human observers, this process is limited by subjectivity, high costs, and lack of scalability. Although deep learning models, particularly convolutional neural networks (CNNs), have been employed for automated cloud analysis, relatively few studies have addressed hazardous convective cloud classification from ground-based sky imagery, and vision transformer architectures have yet to be thoroughly evaluated for this purpose. This research addresses that gap by assessing several transformer-based vision architectures on both daytime all-sky camera images and satellite imagery. The results show that transformer-based vision models outperform CNN baselines for both modalities, and achieving a Probability of Detection above 90% and False Alarm Ratios below 20% for ground-based camera imagery, thus meeting operational thresholds defined by meteorological authorities. These findings demonstrate the potential of using vision transformers in improving automated weather monitoring systems, thereby contributing to enhanced operation safety in aviation.