Wyniki wyszukiwania - BazTech

1

Recommendations for planning UAV flight missions for geodata collection

Hutsul Taras, Lysko Bogdan, Zhezhera Ivan, Tkach Vladyslav, Tsvyk Tetyana

Reports on Geodesy and Geoinformatics

|

2025

|

Vol. 119

62--70

EN

A number of different types of information are generally associated with places. It is estimated that about 75-90 % of information may contain an official link to a specific area, expressed as, for example, coordinates, or addresses, and therefore has a spatial character, making data collection a responsible and important stage, which reasonably affects the quality of its results. Information and its sources are treated with particular care and rigor in the scientific field: in most cases, the data must be relevant, reliable, technically simple, and collected quickly at reasonable costs. The analysis of geographic information makes it possible to obtain qualitatively new information and reveal previously unknown patterns. Modern data collection methods are divided into three distinct groups: terrestrial, cartographic, and remote. Remote or aerospace methods are considered to be those that allow information to be collected. It refers to objects on the Earth's surface, phenomena, or processes from space or the atmosphere, recorded by detecting electromagnetic radiation on the ground across various spectral ranges. The involvement of various platforms (providers) of surveillance equipment makes it possible to divide them into: space, aerial photography, and images from Unmanned Aerial Vehicles (UAVs). As a technology justified on security grounds, UAVs show great promise in many areas of application. Effective planning of drone missions allows for the collection of larger sets of data with a higher level of detail and in a shorter period of time. The continuity of information collection for a given territory allows for the most accurate and reliable three-dimensional modelling, spatial analysis and geostatistics of the local situation.

2

Hoofed animal detection in UAV thermal images using Balanced Random Forest and CNN features

Włodarczyk Dorota, Jóźków Grzegorz

Reports on Geodesy and Geoinformatics

|

2025

|

Vol. 120

1--13

PL

Wildlife monitoring is vital to conservation efforts and the prevention of animal-related negative impacts on human activities and ecosystems. The use of Unmanned Aerial Vehicles (UAVs) enables data collection with no harm to wildlife and in difficult field conditions. This study proposes a method of detecting hoofed animals in UAV-acquired thermal images, addressing the challenges of low-resolution thermal imaging and the presence of other heated objects hindering simple temperature analysis and image segmentation. The proposed method uses machine learning algorithms and is designed to work with a limited size of training dataset. The method consists of an initial segmentation step that detects potential animals based on thermal and geometrical signatures, followed by classification using a Balanced Random Forest (BRF) algorithm. One of the key aspects of the proposed method is the use of geometric and thermal features along with multi-scale Convolutional Neural Network (CNN) extracted feature representations in BRF. The benefit of the BRF is its speed, little requirement regarding the amount of training data, and its capacity to work with an imbalanced number of objects in different classes. The dataset was collected during two UAV flights over a fenced enclosure with wild hoofed animals. The proposed approach showed high efficiency, achieving an overall accuracy of 90%. These results confirm the feasibility of UAV-based animal detection based solely on thermal images collected during the day and showing many other heated objects. The method provides a solution for wildlife monitoring, with potential adaptability to different species and further applications.

3

Spatiotemporal variation of vegetation cover in mining areas of YSR Kadapa district, Andhra Pradesh using remote sensing and GIS

Yarlanki Vinod Kumar, Muppala Prudhvi R., Yasam Prem K., Gollapudi Balaji, Moddu Sumanth, Besta Appanna

Journal of Sustainable Mining

|

2025

|

Vol. 24, iss. 1

102--116

EN

The spatiotemporal variation of vegetation cover in the mining areas of YSR Kadapa district, Andhra Pradesh, using Remote Sensing and GIS techniques. By focusing on the years 2014, 2018, and 2023, the analysis provides insights into changes observed in water bodies, bare soil, sparse vegetation, moderate vegetation, and dense vegetation. The results reveal dynamic trends in land cover categories, highlighting the environmental impact of mining activities in the region. A significant decline in water bodies is observed, with the area reducing from 3.48 km2 in 2014 to 1.91km2 in 2023. This decrease raises concerns about the potential degradation of aquatic ecosystems, reflecting the ecological consequences of mining operations. The fluctuating pattern in bare soil areas, increasing from 37.64 km2 in 2014 to 40.37 km2 in 2018 and subsequently decreasing to 34.65 km2 in 2023, indicates the complex nature of land use changes and reclamation efforts in the mining regions. The study highlights a significant decrease in sparse vegetation from 6.88 km2 in 2014 to 4.43 km2 in 2018, followed by a substantial increase to 13.49 km2 in 2023. This suggests the resilience of vegetation in certain areas or potential reforestation initiatives. A consistent decline in moderate vegetation is observed, with the area decreasing from 5.72 km2 in 2014 to 4.25 km2 in 2023, indicating the lasting impacts of mining on plant health and ecosystem stability. Fluctuations in dense vegetation areas are noted, with a decrease from 2.31 km2 in 2014 to 1.72 km2 in 2023. This decline may signify habitat disruption and environmental stress resulting from mining operations. The consequences of these spatiotemporal changes in vegetation cover extend beyond the immediate landscape, impacting the ecosystem and environment. The reduction in water bodies and vegetation, coupled with an increase in bare soil, suggests potential biodiversity loss, soil erosion, and altered hydrological patterns. These changes pose significant challenges, affecting local fauna and flora and contributing to broader ecological imbalances. The study emphasizes the importance of employing sustainable mining practices to mitigate these adverse effects, ensuring the long-term environmental health and resilience of the region. Sustainable practices could include measures to protect and restore water bodies, prevent soil erosion, and promote reforestation and habitat conservation. By adopting such practices, the mining industry can help preserve biodiversity, maintain ecosystem services, and support the overall environmental sustainability of the YSR Kadapa district.

4

Remote sensing and GIS based crop monitoring: a case study of Tavra village in Vadodara

Zalavadiya Umangkumar B., Sabhadiya Himanshi R., Kapupara Parth J.

GIS Odyssey Journal

|

2025

|

Vol. 5, no. 1

167--175

EN

Food security is increasingly challenged by environmental changes, natural resource degradation, and population growth. Crop yields have already stagnated in many regions and are further affected by rising temperatures. The growing global population imposes a direct demand on agriculture to produce food, fiber, and fodder, necessitating the consumption of vast amounts of water. To maximize agricultural productivity and ensure sustainable crop yields, continuous crop monitoring is essential. Remote sensing has emerged as a powerful technology for vegetation monitoring, enabling spectral analysis of high-resolution satellite imagery to assess crop health and development. This study utilizes remote sensing techniques in conjunction with Geographic Information Systems (GIS) to monitor crop conditions. The Green Chromatic Coordinate (GCC) and Normalized Difference Vegetation Index (NDVI) were estimated using Landsat-9 satellite imagery. The analysis was conducted using QGIS for Tavra Village Farm, near Parul University, Waghodia, Vadodara, Gujarat, India. The observed GCC values ranged from 0.9352 to 0.3297, while NDVI values varied between 0.3300 and 0.0398 over the temporal period. The trend analysis of GCC and NDVI indicated an initial increase from November (early crop growth stage) to January (mid-growth stage), followed by a decline by February (crop maturity stage). These findings demonstrate the effectiveness of remote sensing and GIS in monitoring crop growth patterns, offering valuable insights for precision agriculture and resource management.

5

Climate-induced dynamics of ice cover in southeastern Greenland (2000–2024) revealed by satellite remote sensing

Głowienka Ewa, Mizera Mikołaj

Geomatics, Landmanagement and Landscape

|

2025

|

no. 3

71--81

EN

This study investigates multidecadal changes in the extent of glacierized surfaces in Kommune Kujalleq, southeastern Greenland, between 2000 and 2024. Multispectral satellite imagery from Landsat 7 and Landsat 8, combined with climate data from MODIS and the Global Precipitation Measurement (GPM) mission, was used to quantify changes in summer ice cover and to evaluate their relationship with atmospheric drivers. The Normalized Difference Snow Index (NDSI ≥ 0.4) was applied to classify ice-covered pixels during the melt season (July–September), and climate variables were derived for both summer and winter seasons. The results reveal an overall net decline in ice-covered area of approximately 4% (about 1,600 km²) over the 24-year period, with substantial interannual variability. Years such as 2015 and 2020 exhibited temporary increases in ice extent, coinciding with anomalously high snowfall and below-average summer temperatures, whereas significant losses occurred during warm and dry periods, notably in 2010 and 2024. Despite these fluctuations, the general trend remains one of retreat, driven primarily by sustained Arctic warming. The study highlights the effectiveness of remote sensing and cloud-based geospatial platforms for long-term cryospheric monitoring and contributes to a better understanding of regional glacier sensitivity to climatic variability in the context of global sea-level rise.

PL

W niniejszym badaniu analizowano zmiany zasięgu powierzchni zlodowaconych w Kommune Kujalleq, w południowo-wschodniej Grenlandii, w okresie wielu dekad, w latach 2000–2024. Wielospektralne zdjęcia satelitarne z satelitów Landsat 7 i Landsat 8, w połączeniu z danymi klimatycznymi z MODIS i misji Global Precipitation Measurement (GPM), wykorzystano do ilościowego określenia zmian letniej pokrywy lodowej i oceny ich związku z czynnikami atmosferycznymi. Do klasyfikacji pikseli pokrytych lodem w sezonie topnienia (lipiec–wrzesień) zastosowano znormalizowany wskaźnik różnicy śniegu (NDSI ≥ 0,4), a zmienne klimatyczne wyznaczono zarówno dla sezonu letniego, jak i zimowego. Wyniki wskazują na ogólny spadek netto powierzchni pokrytej lodem o około 4% (około 1600 km²) w ciągu 24 lat, ze znaczną zmiennością międzyroczną. Lata takie jak 2015 i 2020 charakteryzowały się tymczasowym wzrostem zasięgu lodu, zbiegającym się z anomalnie wysokimi opadami śniegu i temperaturami lata poniżej średniej, natomiast znaczne straty wystąpiły w okresach ciepłych i suchych, zwłaszcza w 2010 i 2024 roku. Pomimo tych wahań, ogólna tendencja nadal zmierza w kierunku cofania się lodu, napędzanego głównie przez utrzymujące się ocieplenie Arktyki. Badanie podkreśla skuteczność teledetekcji i chmurowych platform geoprzestrzennych do długoterminowego monitorowania kriosfery i przyczynia się do lepszego zrozumienia regionalnej wrażliwości lodowców na zmienność klimatu w kontekście globalnego wzrostu poziomu morza.

6

Implementation of Object Detection Algorithms in the Inventory of Technical Infrastructure, Case Study of Utility Poles Detection

Czyrzniak Mateusz, Rapiński Jacek

Civil and Environmental Engineering Reports

|

2025

|

Vol. 35, no. 1

177--189

EN

Roads, bridges, railways, poles, and power lines are one of main elements of technical infrastructure. Their proper condition is essential for ensuring people, goods, and electricity transportation. Inventory is a crucial process for maintaining the health of technical infrastructure. Conducting inventory with traditional surveying techniques, such as Global Navigation Satellite Systems (GNSS) or trigonometry, is both time and money consuming. In recent years, modern remote sensing techniques like LiDAR (Light Detection and Ranging) and aerial and ground based imaging have been employed for inventory purposes. These methods enable data collection over large areas, reducing both the time and cost of measurement processes. Data from these devices is often processed in conjunction with artificial intelligence algorithms that assist in processing large data sets. This paper aims to introduce deep learning (DL) algorithms for inventory purposes using object detection with utility poles as a case study. The research was conducted using empirical data. A key element in training a DL model is the optimal selection of hyperparameters. During the study, ten models were trained, and their performance was compared based on selected metrics. For this study, a dataset of 1,736 original utility pole images was used. To augment the data, new images were created through rotation and mirroring. The best model achieved the following results: Precision = 98.82%, Recall = 97.29%, F1-score = 98.05%, mAP50 = 97.92%, mAP75 = 89.93%. The performance of the model gives a solid base for further implementation of DL object detection techniques for inventory of technical infrastructure.

7

Integrated environmental impact assessment of the Sasa tailing dam failure using remote sensing techniques

Adjiski Vancho, Zubíček Václav

Archives of Mining Sciences

|

2025

|

Vol. 70, no. 3

525--556

EN

Failures of tailings dams represent a critical environmental hazard, releasing mining by-products that cause long-term damage to nearby ecosystems. This research presents a detailed analysis using remote sensing techniques of the 2003 Sasa tailing dam disaster in North Macedonia. By utilising Landsat 5 imagery and Google Earth Engine (GEE), multiple spectral indices — including the Normalised Difference Vegetation Index (NDVI), Normalised Difference Moisture Index (NDMI), Normalised Difference Water Index (NDWI), Modified NDWI (MNDWI), and a turbidity proxy — were integrated to examine the immediate and spatial impacts on vegetation health, soil moisture, water presence, and sediment levels. The findings indicated significant ecological fluctuations along the river path, including vegetation stress, changes in soil moisture, water pooling, and turbidity. These effects displayed spatial gradients, diminishing further from the contamination pathway and forming distinct zones of influence. Certain intermediate areas showed anomalous disturbances, where sediment and hydrological changes impeded vegetation recovery. Pixel-level, buffer-based, and zone-based analyses - combined with Z-scores and correlation studies — revealed a complex post-disaster landscape. Weak correlation with topographic features suggested that localised conditions, rather than large-scale gradients, governed short-term ecological recovery. The study provides a framework for integrated, multi-index and multiscale environmental impact assessment, contributing to improved remediation strategies, disaster response planning, and sustainable management of post-mining landscapes.

8

Advancing flood disaster management: leveraging deep learning and remote sensing technologies

Roohi Mohammad, Reza Ghafouri Hamid, Mohammad Ashrafi Seyed

Acta Geophysica

|

2025

|

Vol. 73, no. 1

557--575

EN

Floods are among the most widespread and devastating natural disasters, accounting for 47% of all weather-related events and affecting over 2.3 billion people, particularly in Asia. Assessing flood-prone areas is crucial for effective disaster risk reduction, but existing flood damage estimation methods, such as depth-damage functions, often lack regional adaptability and accuracy. This study addresses this gap by integrating geospatial data, remote sensing, and artificial intelligence (AI) to identify flood-affected areas in the Kan basin, Tehran. We applied deep learning methods, specifically U-Net and fully convolutional neural network (FCN) algorithms, to optical and radar images from four flood events. Our results demonstrate that the U-Net model achieves significantly higher accuracy (88%) in identifying flood-affected areas compared to the FCN model (55% accuracy). This superior performance is further supported by the mean intersection over union (mIoU) values, with U-Net achieving 0.65, compared to 0.55 for FCN. The key message of this investigation is that deep learning, particularly the U-Net model, applied to remote sensing data holds significant promise for enhancing flood monitoring, early warning systems, and disaster management strategies by enabling more accurate and timely flood assessments.

9

Analysis of the Effectiveness of Selected Machine Learning Algorithms in the Classification of Satellite Image Content Depending on the Size of the Training Sample

Kupidura Przemysław, Niemyski Stanisław

Teledetekcja Środowiska

|

2024

|

T. 64

24--38

EN

The article presents an analysis of the accuracy of 3 popular machine learning (ML) methods: Maximum Likelihood Classifier (MLC), Support Vector Machine (SVM), and Random Forest (RF) depending on the size of the training sample. The analysis involved performing the classification of the content of a Landsat 8 satellite image (divided into 6 basic land cover classes) in 10 different variants of the number of training samples (from 2664 to 34711 pixels), estimating individual results, and a comparative analysis of the obtained results. For each classification variant, an error matrix was developed and on their basis, accuracy metrics were calculated: f1-score, precision and recall (for individual classes) as well as overall accuracy and kappa index of agreement (generally for the entire classification). The analysis showed a stimulating effect of the size of the training sample on the accuracy of the obtained classification results in all analyzed cases, with the most sensitive to this factor being MLC, showing the best effectiveness with the largest training sample and the smallest - with the smallest, and the least SVM, characterized by the highest accuracy with the smallest training sample, comparing to other algorithms.

PL

Artykuł przedstawia analizę dokładności 3 popularnych metod uczenia maszynowego: Maximum Likelihood Classifier (MLC), Support Vector Machine (SVM) oraz Random Forest (RF) w zależności od liczebności próbki treningowej. Analiza polegała na wykonaniu klasyfikacji treści zdjęcia satelitarnego Landsat 8 (w podziale na 6 podstawowych klas pokrycia terenu) w 10 różnych wariantach liczebności próbek uczących (od 2664 do 34711 pikseli), oszacowaniu poszczególnych wyników oraz analizie porównawczej uzyskanych wyników. Dla każdego wariantu klasyfikacji opracowano macierz błędów, a na ich podstawie obliczono metryki dokładności: F1-score, precision and recall (dla pojedynczych klas) oraz ogólną dokładność i wskaźnik zgodności Kappa (ogólnie dla całej klasyfikacji). Analiza wykazała stymulujący wpływ rozmiaru próbki uczącej na dokładność uzyskiwanych wyników klasyfikacji we wszystkich analizowanych przypadkach, przy czym najbardziej wrażliwym na ten czynnik był MLC, wykazujący się najlepszą skutecznością przy największej próbce treningowej i najmniejszą - przy najmniejszej, a najmniej SVM, cechujący się największą dokładnością przy najmniejszej próbce treningowej, w porównaniu do pozostałych algorytmów.

10

Assessing the impact of land cover and land use change on urban infrastructure resilience in Abuja, Nigeria: a case study from 2017 to 2022

Adewumi Rowland, Agbasi Okechukwu Ebuka

Structure and Environment

|

2024

|

Vol. 16, no. 1

6--17

EN

The remarkable feature of rapid urbanisation, which has fundamentally altered the distribution of land cover and land use (LULC), is what sets the contemporary era apart. The impact of these modifications on the resilience of Abuja’s metropolitan infrastructure from 2017 to 2022 is examined in this study. Our study examined the dynamic changes in LULC using information from remote sensing, geospatial analysis software, and land cover categorization techniques. The findings indicate major changes in Abuja’s topography, including a decrease in the number of water bodies, a decrease in the number of trees, the expansion of urban areas, changes in agricultural land use, and fluctuations in the amount of grazing land. The previously mentioned changes have significant consequences for urban infrastructure resilience, affecting various sectors such as water supply, transportation, housing, utilities, and food distribution systems. The infrastructure supporting water supply and sanitation may be severely stretched as the number of water bodies decreases, affecting the quantity and quality of accessible water supplies. As metropolitan areas expand, greater strain is placed on transportation infrastructure, exacerbating traffic congestion and complicating road maintenance difficulties. Changes in agricultural land use can have an impact on food production and distribution, hence affecting food security. Deforestation can cause ecological deterioration, affecting a variety of aspects such as temperature regulation, biological diversity, and atmospheric purity. Adaptive approaches, green infrastructure, and adopting sustainable urban design can all strengthen the resilience of urban infrastructure, according to this study. The significance of renewable energy adoption, community participation, green building laws, the establishment of public-private partnerships, integrated water resource management, and data-driven decision-making is emphasised. Improving legal frameworks that prioritise resilience and sustainability is critical. It is critical to have a complete grasp of the complicated links between changes in LULC, and the resilience of urban infrastructure in order to enable educated urban design and decision-making processes. Policymakers and urban planners may address and minimise the negative consequences of climate change while improving the overall quality of life in cities by using sustainable development practises. The findings of this study have the potential to dramatically improve Abuja’s people’s well-being and sustainability, especially given the variety of urban difficulties they encounter.

PL

Współczesną erę wyróżnia niezwykle szybka urbanizacja, która zasadniczo zmieniła rozkład pokrycia terenu i użytkowania gruntów (LULC). W niniejszym badaniu przeanalizowano wpływ tych zmian na odporność infrastruktury metropolitalnej Abudży w latach 2017-2022. Dynamiczne zmiany LULC zbadano przy użyciu informacji z teledetekcji, oprogramowania do analizy geoprzestrzennej oraz technik kategoryzacji pokrycia terenu. Wyniki wskazują na poważne zmiany w topografii Abudży, w tym spadek liczby zbiorników wodnych, spadek liczby drzew, ekspansję obszarów miejskich, zmiany w użytkowaniu gruntów rolnych i wahania w ilości pastwisk. Zmiany te mają znaczące konsekwencje dla odporności infrastruktury miejskiej, wpływając na różne sektory, takie jak zaopatrzenie w wodę, transport, mieszkalnictwo, usługi komunalne i systemy dystrybucji żywności. Infrastruktura wspierająca zaopatrzenie w wodę i urządzenia sanitarne może być poważnie obciążona, ponieważ malejąca liczba zbiorników wodnych odbija się na ilości i jakości dostępnych zasobów wody. Wraz z rozwojem obszarów metropolitalnych rośnie obciążenie infrastruktury transportowej, co zwiększa natężenie ruchu i komplikuje utrzymanie dróg. Zmiany w użytkowaniu gruntów rolnych wpływają na produkcję i dystrybucję żywności, a tym samym na bezpieczeństwo żywnościowe. Wylesianie może powodować pogorszenie stanu środowiska, wpływając na regulację temperatury, różnorodność biologiczną i czystość atmosfery. Według naszych badań podejście adaptacyjne, zielona infrastruktura i przyjęcie zrównoważonego projektowania urbanistycznego mogą wzmocnić odporność infrastruktury miejskiej. Podkreśla się znaczenie energii odnawialnej, udziału społeczności, przepisów dotyczących zielonego budownictwa, ustanowienia partnerstw publiczno-prywatnych, zintegrowanego zarządzania zasobami wodnymi i podejmowania decyzji w oparciu o dane. Kluczowe znaczenie ma poprawa ram prawnych, które powinny priorytetowo traktować kwestie odporności miejskiej oraz zrównoważonego rozwoju. Świadome projektowanie urbanistyczne i procesy decyzyjne możliwe są jedynie przy zrozumieniu skomplikowanych powiązań między zmianami w LULC a odpornością infrastruktury miejskiej. Zastosowanie praktyk zrównoważonego rozwoju umożliwi decydentom i urbanistom zminimalizowanie negatywnych konsekwencji zmian klimatycznych oraz podniesienie ogólnej jakości życia w miastach. Wyniki tego badania mogą potencjalnie znacznie poprawić dobrobyt i zrównoważony rozwój mieszkańców Abudży, zwłaszcza biorąc pod uwagę różnorodność napotykanych przez nich trudności miejskich.

11

Sat4BIM4D - the concept of using satellite remote sensing to monitor construction progress in conjunction with BIM

Glinka Szymon

Reports on Geodesy and Geoinformatics

|

2024

|

Vol. 118

142--154

EN

Monitoring the progress of construction work and adhering to the schedule is crucial for the timely completion of projects. Integrating data from various sensors (e.g., cameras, laser scanners) mounted on diverse platforms (rovers, drones, satellites) with BIM 4D (Building Information Modelling) enables effective construction control solutions. By leveraging 3D models enriched with temporal information, project management can be significantly enhanced. This paper focuses on a comprehensive review of current literature and state-of-the-art practices to design a framework for integrating satellite remote sensing data with BIM 4D, termed the Sat4BIM4D method. Proposals for this method are developed alongside algorithms for processing satellite-derived data to monitor construction progress, particularly for infrastructure projects. The study emphasizes the compatibility and synergy between satellite data and BIM 4D, providing a structured direction for future research. Advantages, limitations, and potential challenges of the proposed approach are also critically analyzed to pave the way for further development in this domain.

12

Geoprocessing of archival aerial photos and their scientific applications: A review

Kostrzewa Adam

Reports on Geodesy and Geoinformatics

|

2024

|

Vol. 118

1--16

EN

Poland as well as other countries keep extensive collections of 20th and 21st-century aerial photos, which are underexploited compared to such other archival materials as satellite imagery. Meanwhile, they offer significant research potential in various areas, including urban development, land use changes, and long-term environmental monitoring. Archival photographs are detailed, often obtained every five to ten years, and feature high resolution, from 20 cm to 1 m. Their overlap can facilitate creating precise digital models that illustrate topography and land cover, which are essential variables in many scientific contexts. However, rapidly transforming these photographs into geographically accurate measurements of the Earth’s surface poses challenges. This article explores the obstacles in automating the processing of historical photographs and presents the main scientific research directions associated with these images. Recent advancements in enhancing workflows, including the development of modern digital photogrammetry tools, algorithms, and machine learning techniques are also discussed. These developments are crucial for unlocking the full potential of aerial photographs, making them easier accessible and valuable for a broader range of scientific fields. These underutilized photographs are increasingly recognized as vital in various research domains due to technological advancements. Integrating new methods with these historical images offers unprecedented opportunities for scientific discovery and historical understanding, bridging the past with the future through innovative research techniques.

13

Comparative analysis of the performance of selected machine learning algorithms depending on the size of the training sample

Kupidura Przemysław, Kępa Agnieszka, Krawczyk Piotr

Reports on Geodesy and Geoinformatics

|

2024

|

Vol. 118

53--69

EN

The article presents an analysis of the effectiveness of selected machine learning methods: Random Forest (RF), Extreme Gradient Boosting (XGB), and Support Vector Machine (SVM) in the classification of land use and cover in satellite images. Several variants of each algorithm were tested, adopting different parameters typical for each of them. Each variant was classified multiple (20) times, using training samples of different sizes: from 100 pixels to 200,000 pixels. The tests were conducted independently on 3 Sentinel-2 satellite images, identifying 5 basic land cover classes: built-up areas, soil, forest, water, and low vegetation. Typical metrics were used for the accuracy assessment: Cohen's kappa coefficient, overall accuracy (for whole images), as well as F-1 score, precision, and recall (for individual classes). The results obtained for different images were consistent and clearly indicated an increase in classification accuracy with the increase in the size of the training sample. They also showed that among the tested algorithms, the XGB algorithm is the most sensitive to the size of the training sample, while the least sensitive is SVM, which achieved relatively good results even when using training samples of the smallest sizes. At the same time, it was pointed out that while in the case of RF and XGB algorithms the differences between the tested variants were slight, the effectiveness of SVM was very much dependent on the gamma parameter -- with too high values of this parameter, the model showed a tendency to overfit, which did not allow for satisfactory results.

14

Teledetekcja wymaga wsparcia naziemnymi systemami obserwacyjnymi: przypadek polsko-tureckiego projektu - Silva Nympha

Bęcek Kazimierz

Przegląd Geodezyjny

|

2024

|

R. 96, nr 10

15--17

PL

W pracy przedstawiono podstawowe informacje o projekcie Silva Nympha pt. „Zrównoważone użytkowanie i inteligentne zarządzanie lasami”, który zostały wyselekcjonowany do finansowania w konkursie Narodowego Centrum Badań i Rozwoju (NCBR) w ramach bilateralnej współpracy pomiędzy Polską i Turcją w zakresie badań naukowych. Omówiono organizację projektu, podkreślono konieczność łączenia danych teledetekcyjnych z pomiarami i obserwacjami naziemnymi dla uzyskania poprawnych i użytecznych opracowań. Przedstawiono model łączenia danych w postaci hiperkostki, która tworzona jest dla każdego piksela obiektu badawczego wzdłuż osi czasu. Zawartość hiperkostek, reprezentujących obszar badawczy, będzie przedmiotem studiów metodami sztucznej inteligencji. W pracy opisano składowe systemu obserwacyjnego zainstalowanego na potrzeby projektu na polu badawczym.

EN

The paper presents basic information about the Silva Nympha project entitled "Sustainable use and intelligent management of forests", which was selected for funding in the competition of the National Center for Research and Development (NCBR) within the framework of bilateral cooperation between Poland and Turkey in the field of scientific research. The organization of the project was discussed, and the need to combine remote sensing data with measurements and ground observations to obtain correct and valid results was emphasized. A model for integrating remote sensing and in situ data was proposed as a hypercube, created for each pixel of the research object along the time axis. The content of hypercubes, representing the research area, will be the subject of studies using artificial intelligence methods. The paper describes the components of the observation system installed to meet the project's needs in the research field.

15

Wykorzystanie teledetekcji i uczenia maszynowego do oceny stanu terenów zielonych

Radkowski Adam, Kluwak Konrad, Rusiecki Andrzej, Ciskowski Piotr, Jeleń Łukasz, Wiącek Paweł

Przegląd Budowlany

|

2024

|

R. 95, nr 5

187--191

PL

Prezentowane badania skupiają się na wykorzystaniu technik teledetekcyjnych w analizie stanu terenów zielonych. W pierwszej części badań główny nacisk kładziony jest na ocenę wizualną, wskaźniki wegetacyjne oraz chemiczną analizę próbek roślin. Za pomocą kamery multispektralnej i RGB pozyskuje się wielokanałowe obrazy, służące do wnioskowania na temat zdrowia roślin, ich kondycji oraz zawartości różnych składników chemicznych. Do tego celu użyto głębokich sieci neuronowych i zaawansowanych metod uczenia maszynowego. Praca prezentuje pionierskie narzędzie do monitorowania i zarządzania zielonymi przestrzeniami w duchu zrównoważonego rozwoju.

EN

The presented research focuses on the use of remote sensing techniques in the analysis of the condition of green areas. In the first part of the research, the main emphasis is placed on visual assessment, vegetation indicators and chemical analysis of plant samples. Using a multispectral and RGB camera, multi-channel images are obtained, used to draw conclusions about the health of plants, their condition and the content of various chemical components. For this purpose, deep neural networks and advanced machine learning methods were used. The work presents a pioneering tool for monitoring and managing green spaces in the spirit of sustainable development.

16

Sub-Micrometer Particles Remote Detection in Enceladus’ Plume Based on Cassini’s UV Spectrograph Data

Kotlarz Jan, Kubiak Katarzyna, Zalewska Natalia

Pomiary Automatyka Robotyka

|

2024

|

R. 28, nr 1

107--111

EN

Enceladus is the Saturnian satellite is known to have water vapor erupting from its south pole region called „Tiger Stripes”. Data collected by Cassini Ultraviolet Imaging Spectrograph during Enceladus transiting Saturn allow us to estimate water plume absorption from 1115.35-1912.50 Å and compare it to the Mie solutions of Maxwell equations for particles with a diameter in the range from 10 nm up to 2 μm. The best fit performed using Gradient Descent method indicates a presence of sub-micrometer particles of diameters: 120-180 nm and 240-320 nm consistent with Thermofilum sp., Thermoproteus sp., and Pyrobaculum sp. cell sizes present in hydrothermal vents on Earth.

PL

Enceladus, księżyc Saturna, jest charakterystyczny ze względu na erupcje pary wodnej z regionu jego bieguna południowego, tzw. „Tiger Stripes”. Dane zebrane przez instrument sondy Cassini: Ultraviolet Imaging Spectrograph podczas tranzytu Enceladusa przed tarczą Saturna pozwalają oszacować absorpcję światła przez pióropusze wodne w zakresie 1115,35-1912,50 Å i porównać ją z rozwiązaniami Mie równań Maxwella dla cząsteczek o średnicach w zakresie od 10 nm do 2 μm. Najlepsze dopasowanie wykonane metodą Gradient Descent wskazuje na obecność cząstek sub-mikrometrowych o średnicach: 120-180 nm i 240-320 nm zgodnych z rozmiarami komórek Thermofilum sp., Thermoproteus sp. i Pyrobaculum sp. obecnych w kominach hydrotermalnych na Ziemi.

17

Monitoring Coastal Erosion and Deposition in Sam Son City, Vietnam – a Contribution from Remote Sensing Data

Pham Lan Thi, Le Dung KLim, Tong Son Si, Le Ha Thi Thu

Inżynieria Mineralna

|

2024

|

Vol. 2, no. 1

79--87

EN

Studying the trends in shoreline erosion and accretion is essential for a wide range of investigations conducted by coastal scientists, and coastal managers. Shoreline erosion and accretion occur as a result of both natural and human influences. Some areas along shoreline in Sam Son are eroded and deposed by natural coastal processes and human actions, such as storm, wave, tourism activities. Purpose of this work is to study the erosion and deposition in Sam Son over 33 years (1989–2022). Coastlines were extracted using multi-temporal Landsat images, and the shoreline change rate was determined using Digital Shoreline Analysis Systems (DSAS). The results of this paper inlustrated that the shoreline change in Sam Son undergoes significant and varied fluctuations across different areas. At the Hoi estuary, erosion rates vary from -2.22 m/year to -40.32 m/year. The construction of FLC Sam Son is one of the factors contributing to sedimentation loss in the northern part of Sam Son City, which is situated adjacent to the East Sea and next to the Ma River. Furthermore, the accretion rate has strongly increased, reaching 9.7 m/year in the Do River estuary. The phenomenon of sediment deposition serves as the basic for constructing hotels to cater to tourism in Sam Son.

PL

Badanie trendów erozji i akrecji linii brzegowej jest istotne dla szerokiego zakresu badań prowadzonych przez naukowców nadmorskich i menedżerów wybrzeża. Erozja i akrecja linii brzegowej występują zarówno w wyniku procesów naturalnych, jak i wpływu człowieka. Pewne obszary wzdłuż linii brzegowej w Sam Son ulegają erozji i akrecji w wyniku naturalnych procesów przybrzeżnych i działań ludz-kich, takich jak burze, fale, działalność turystyczna... Celem tej pracy jest zbadanie erozji i akrecji w Sam Son na przestrzeni 33 lat (1989–2022). Linie brzegowe zostały wyodrębnione z wykorzystaniem wieloczasowych obrazów satelitarnych Landsat, a wskaźnik zmian linii brzegowej został określony przy użyciu systemów cyfrowej analizy linii brzegowej (DSAS). Wyniki tej pracy ilustrują, że zmiany linii brze-gowej w Sam Son podlegają znacznym i zróżnicowanym fluktuacjom w różnych obszarach. W ujściu rzeki Hoi wskaźniki erozji wahają się od -2,22 m/rok do -40,32 m/rok. Budowa FLC Sam Son jest jednym z czynników przyczyniających się do utraty osadów w północnej części miasta Sam Son, która sąsiaduje z Morzem Wschodnim i rzeką Ma. Ponadto, wskaźnik akrecji znacząco wzrósł, osiągając 9,7 m/rok w ujściu rzeki Do. Zjawisko osadzania się osadów służy jako podstawa do budowy hoteli obsługujących turystykę w Sam Son

18

Identification of salt-affected soils in the coastal area Of Krishna District, Andhra Pradesh, using remote sensing data and machine learning techniques

Anuradha Govada, Chivukula Venkata Sai Sankara Vineeth, Kothagundla Naga Ganesh

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2024

|

T. 14, nr 2

83--88

EN

In agricultural soil analysis, the challenge of soil salinization in regions like Krishna District, Andhra Pradesh, profoundly impacts soil health, crop yield, and land usability, affecting approximately 77,598 hectares of land. To address this issue, three machine learning algorithms are compared for classifying salinity levels in the coastal area of Krishna district, Machilipatnam. This study utilizes Landsat-8 images from 2014 to 2021, correcting for cloud cover and creating a true-color composite. The study area is defined and visualized. Twelve indices, derived from Landsat imagery, are incorporated into the analysis. These indices, including spectral bands and mathematical expressions, are added as image bands. The median of these indices is calculated, and sample points representing both non-saline and saline areas are used for supervised machine learning. The data is divided into two sets: training and validation. The study evaluates Random Forest, Classification and Regression Trees, and Support Vector Machines for classifying soil salinity levels using these indices. The RF algorithm produced an accuracy of 92.1%, CART produced 91.3%, and SVM produced 86%. Results are displayed on the map, representing predicted salinity levels with distinct colors. Performance metrics are evaluated, and they assess algorithm performance. The research involved gives insights into the classification of soil salinity using machine learning, which could represent an efficient solution to the problem of soil salinization in Machilipatnam.

PL

W rolniczej analizie gleby, wyzwanie zasolenia gleby w regionach takich jak dystrykt Krishna, Andhra Pradesh, głęboko wpływa na zdrowie gleby, plony i użyteczność gruntów, wpływając na około 77 598 hektarów ziemi. Aby rozwiązać tę kwestię, porównano trzy algorytmy uczenia maszynowego do klasyfikacji poziomów zasolenia w obszarze przybrzeżnym dystryktu Krishna, Machilipatnam. W badaniu wykorzystano obrazy Landsat-8 z lat 2014-2021, korygując je pod kątem zachmurzenia i tworząc kompozycję w prawdziwych kolorach. Obszar badań został zdefiniowany i zwizualizowany. Do analizy włączono dwanaście wskaźników pochodzących ze zdjęć Landsat. Wskaźniki te, w tym pasma widmowe i wyrażenia matematyczne, są dodawane jako pasma obrazu. Mediana tych wskaźników jest obliczana, a przykładowe punkty reprezentujące zarówno obszary niezasolone, jak i zasolone są wykorzystywane do nadzorowanego uczenia maszynowego. Dane są podzielone na dwa zestawy: treningowy i walidacyjny. W badaniu oceniono Random Forest, Classification and Regression Trees i Support Vector Machines pod kątem klasyfikacji poziomów zasolenia gleby przy użyciu tych wskaźników. Algorytm RF uzyskał dokładność 92,1%, CART 91,3%, a SVM 86%. Wyniki są wyświetlane na mapie, przedstawiając przewidywane poziomy zasolenia za pomocą różnych kolorów. Oceniane są wskaźniki wydajności i wydajność algorytmów. Przeprowadzone badania dają wgląd w klasyfikację zasolenia gleby przy użyciu uczenia maszynowego, co może stanowić skuteczne rozwiązanie problemu zasolenia gleby w Machilipatnam.

19

Remote sensing in mining: a brief overview and examples

Musiałek Makary, Maksymowicz Marcin

Górnictwo Odkrywkowe

|

2024

|

R. 65, nr 3

4--14

EN

Remote sensing (RS) has become an essential tool in the mining industry, offering efficient methods for data collection, processing and analysis. This paper provides a brief overview of RS applications in mining, focusing on techniques such as spectroscopy, Synthetic Aperture Radar (SAR), Light Detection and Ranging (LiDAR), and thermal imaging. These technologies support activities including mineral exploration, mine planning, operational monitoring, environmental assessment, and reclamation. RS enhances safety and risk management through techniques like InSAR and UAV photogrammetry, while also facilitating the management of mining waste and monitoring environmental impacts on vegetation, soil, water, and air quality. The integration of RS with Geographic Information Systems (GIS) and machine learning (ML) enables advanced predictive modeling and decision-making, driving sustainability and efficiency in mining operations. The paper highlights chosen case studies and emerging trends, underscoring the transformative potential of RS in the mining industry.

PL

Teledetekcja stała się kluczowym narzędziem w branży górniczej, oferując efektywne metody zbierania i przetwarzania i analizy danych. Niniejszy artykuł przedstawia krótki przegląd zastosowań teledetekcji w górnictwie, koncentrując się na technikach takich jak spektroskopia, dane radarowe (SAR), systemy LiDAR i dane termalne. Technologie te wspierają działania związane z eksploracją złóż, projektowaniem kopalń, monitoringiem operacyjnym, oceną środowiskową i rekultywacją. Teledetekcja poprawia bezpieczeństwo i zarządzanie ryzykiem dzięki technikom takim jak InSAR i fotogrametria, a także umożliwia lepsze zarządzanie odpadami górniczymi oraz monitorowanie wpływu działalności górniczej na roślinność, glebę, wodę i jakość powietrza. Integracja teledetekcji z systemami informacji geograficznej (GIS) i uczeniem maszynowym umożliwia zaawansowane modelowanie predykcyjne i podejmowanie decyzji, przyczyniając się do zrównoważonego rozwoju i efektywności operacji wydobywczych. Artykuł przytacza wybrane studia przypadków oraz nowe trendy, pokazując transformacyjny potencjał teledetekcji w przemyśle wydobywczym.

20

Assessment of the quality of the ecological environment of an area with tourist potential using the remote sensing ecological index: Case of El Kala National Park (Algeria)

Karima Chabbi, Zennir Rabah, Khallef Boubaker

Geomatics, Landmanagement and Landscape

|

2024

|

no. 2

31--52

EN

The assessment of the ecological quality of the environment in El Kala National Park plays an important role in the protection and management of its tourist potential in the face of ecological constraints that have arisen. The present study is based on the use of remote sensing data; its main objective is to analyze the ecological quality in a protected area using the remote sensing ecological index which is based on the calculation of vegetation indices based on Landsat images taken in 2013 and 2023. This observation period shows that the values of drought, temperature, and humidity in the study area increased while the greenness values decreased. The RSEI index was calculated using principal component analysis of the fourth indicators (NDVI, WET, NDBSI, and LST) which made it possible to quantitatively analyze, monitor, and dynamically evaluate changes in the ecological quality of the environment in this park over the past 10 years. The results obtained show that the spatio-temporal distribution of the ecological quality of the environment of the park experienced a downward trend from 2013 to 2023 with a regression rate of -10.16% for the classes of good and excellent quality ecological. This study is considered a reference for the formulation of measures aimed at protecting the quality of the environment in El Kala National Park, and also a database to determine monitoring indicators for sites characterized by significant tourism potential.