The impact of AI-driven Advance Data Processing using GOLDENAI platform in the Golden Eye project

The potential of earth observation (EO) data for soil drainage classification and mapping has been investigated in a series of research projects using various EO data at different scale. First, soil drainage classification models are developed for specific land use types using optical EO data (ASTER, LANDSAT or IKONOS) for the segmentation of the study area. Afterwards, the relationship between soil drainage classes and EO data (radar and optical) is integrated into statistical models (discriminant analysis, multiple regression, and decision tree) using soil survey information (soil map and soil profile description and analysis). In a study conducted in the Bras d’Henri watershed (167 km2) to update and upgrade old soil survey maps, a set of five RADARSAT-1 images, with different SAR configuration, and an ASTER image were used for soil drainage mapping. A supervised maximum likelihood classification was applied on ASTER bands for land use segmentation. Backscattering coefficients from RADARSAT-1 data and spectral indices from ASTER data were evaluated in stepwise procedures using the 1612 soil profiles classified by soil survey experts. Classification accuracy was higher (75-88 %) with the decision tree classifier (DTC) compared to the discriminant analysis classifier (DAC). The soil drainage map produced with the DTC method was more similar to the conventional soil drainage map than the one derived from the DAC method. A study on the digital soil drainage mapping was also conducted in a broader agricultural area (Monteregie, near Montreal, QC) at the regional scale (1:40 000) using archived LANDSAT and RADARSAT-1 images. The usefulness of selected IKONOS images was assessed in mapping soil drainage and related soil properties (surface texture). Finally, the potential for soil drainage classification and mapping of the fully polarimetric RADARSAT-2 images is actually under investigation using various decomposition algorithms. Classification results will be presented.

Millions of consumer sport and fitness wearables (CSFWs) are used worldwide, and millions of datapoints are generated by each device. Moreover, these numbers are rapidly growing, and they contain a heterogeneity of devices, data types, and contexts for data collection. Companies and consumers would benefit from guiding standards on device quality and data formats. To address this growing need, we convened a virtual panel of industry and academic stakeholders, and this manuscript summarizes the outcomes of the discussion. Our objectives were to identify (1) key facilitators of and barriers to participation by CSFW manufacturers in guiding standards and (2) stakeholder priorities. The venues were the Yale Center for Biomedical Data Science Digital Health Monthly Seminar Series (62 participants) and the New England Chapter of the American College of Sports Medicine Annual Meeting (59 participants). In the discussion, stakeholders outlined both facilitators of (e.g., commercial return on investment in device quality, lucrative research partnerships, and transparent and multilevel evaluation of device quality) and barriers (e.g., competitive advantage conflict, lack of flexibility in previously developed devices) to participation in guiding standards. There was general agreement to adopt Keadle et al.'s standard pathway for testing devices (i.e., benchtop, laboratory, field-based, implementation) without consensus on the prioritization of these steps. Overall, there was enthusiasm not to add prescriptive or regulatory steps, but instead create a networking hub that connects companies to consumers and researchers for flexible guidance navigating the heterogeneity, multi-tiered development, dynamicity, and nebulousness of the CSFW field.

Soil salinity limits plant growth, reduces crop productivity and degrades soil. Multispectral data from Landsat TM are used to study saline soils in southern Tunisia. This study will explore the potential multivariate statistical analysis, such as principal component analysis (PCA) and cluster analysis to identify the most correlated spectral indices and rapidly predict salt affected soils. Sixty six soil samples were collected for ground truth data in the investigated region. A high correlation was found between electrical conductivity and the spectral indices from near infrared and short-wave infrared spectrum. Different spectral indices were used from spectral bands of Landsat data. Statistical correlation between ground measurements of Electrical Conductivity (EC), spectral indices and Landsat original bands showed that the near and short-wave infrared bands (band 4, band 5 and 7) and the salinity indices (SI 5 and SI 9) have the highest correlation with EC. The use of CA revealed a strong correlation between electrical conductivity EC and spectral indices such abs4, abs5, abs7 and si5. The principal components analysis is conducted by incorporating the reflectance bands and spectral salinity indices from the remote sensing data. The first principal component has large positive associations with bands from the visible domain and salinity indices derived from these bands, while second principal component is strongly correlated with spectral indices from NIR and SWIR. Overall, it was found that the electrical conductivity EC is highly correlated (R2 = -0.72) to the second principal component (PC2), but no correlation is observed between EC and the first principal component (PC1). This suggests that the second component can be used as an explanatory variable for predicting EC. Based on these results and combining the spectral indices (PC2 and abs B4) into a regression analysis, model yielded a relatively high coefficient of determination R2 = 0.62 and a low RMSE = 1.86 dS/m.

To accomplish strategic objectives on zero-pollution, the entire mining life cycle (exploration, extraction, closure, mine- site rehabilitation) needs to develop minimal impact exploration and monitoring technologies and applications which are open to the broadest groups of stakeholders. In this respect Earth Observation (EO), Drone & Proximal Sensing and relevant in-situ data bring significant contribution for both, sustainable management of mineral resources and efficient multi-scale monitoring of mining impacts. In this sense, the purpose of the GEOMIN activity, part of the Group on Earth Observations (GEO) Work Programme [1] [2], is to increase awareness and use of state-of-the-art EO data and methods which represent a novel means for sustainable monitoring and management of mineral resources and efficient multi-scale monitoring mining impacts. How is GoldenEye project through AI-driven tools and applications enhances the GEOMIN community?In the scope of the GoldenEye project, OPT/NET delivered the next generation of AI exploitation system: a hybrid platform which combines the processing & automation capabilities of AI with the natural problem-solving abilities of humans. We have developed dedicated applications with our novel approach based on the Artificial Intelligence Knowledge Packs (AI KPs), integrated in GOLDENAI Engine, to rapidly interpret the geographical patterns and environmental impacts caused by the mining activity.What is the role of the GEO Knowledge Hub (GKH) as the Digital portal in promoting the replicability and re-usability of AI KP in the mining sector and how it relates to Goldeneye ?The GEO Knowledge Hub (GKH) is a central cloud-based digital library providing access to Earth Observations applications developed by GEO. The GEO Knowledge Hub is part of the GEOSS Infrastructure and helps the  GEO to advance Open Knowledge. The scope of the GKH is to promote the replicability and re-usability of EO Applications by sharing with the end users, all the Knowledge Resources essential to fully understand and re-use them. All the Knowledge Resources are directly shared, curated and organized by the Knowledge Provider to ensure replicability with proper documentation.Therefore, several Knowledge Packages (KPs) related to the technological solutions of the GoldenEye project can be found in the GKH. In this paper, the KP related to the GOLDENAI platform will be presented, including the description of the integrated AIKPs, such as:AI KP for mineral mapping - Band-ratios based on WorldView-3  AI KP for mineral mapping - SPCA based on WorldView-3 AI KP for UML clustering based on Copernicus Satellite imagery (Sentinel-1 SLC) The paper has been prepared in the frame of the Horizon 2020 co-funded project GOLDENEYE, which has received funds through the GA 869398. 

Large scale forest inventory plot data are key to monitor forest ecosystems, but while they provide very detailed information at tree level they are limited in resolution in both space and through time. Earth Observation (EO) data offer the opportunity for scaling up plot data and improving the temporal resolution of monitoring. However, there are significant challenges to this, including small field plot sizes, pre-processing and potential GPS errors in aligning the data, whilst the huge amount and variety of EO data introduce substantial challenges of high dimensionality, in addition to the noise of training and testing data, within any AI system. In this work, we fuse plot and Earth Observation data, demonstrating the value of embedding existing and newly EO derived metrics, and selecting the most important features to improve monitoring of forest properties at large scales. In this work we work with Sentinel-1 (SAR) and Sentinel-2 (optical) and inventory data from close to 10,000 plots in Spain, measured from onwards. SAR data require substantial pre-processing due to noise and acquisition, topographic and moisture effects. We used pre-processed SAR data, and filtered for non-shaded slopes, removed plots close to surface water and data collected on days with high precipitation. We masked out clouds from our optical data. After fusing the EO data, we removed disturbed areas using the Global Forest Change Collection and plots with high variability of pixels around them to reduce uncertainty due to the small sizes of the plots. As well as using standard indices (e.g., NDVI, RVI), we derive new metrics of the phenological cycle of the forest from monthly averages of indices and bands by selecting features from peaks and troughs. We reduce dimensionality using principal component analysis and random forest to select the most important features. Chosen features are used for training and evaluating a customized AI system to estimate forest variables such as total basal area, stem density, mean diameter at breast height and forest type. The code implemented in Google Earth Engine JavaScript and Python will be released as open source.

Earth Observation (EO), as a tool to improve efficiency of mineral exploration and mine site monitoring, requires easily accessible robust highly automated data processing algorithms. The Horizon Europe funded Research and Innovation Action project “Multi-source and Multi-scale Earth observation and Novel Machine Learning Methods for Mineral Exploration and Mine Site Monitoring” (MultiMiner, 2023–2036) develops innovative machine learning solutions to support the critical raw material (CRM) independency of EU. We develop and utilize self-supervised or weakly supervised machine learning solutions which require a low number of in situ reference data. This presentation showcases the recent advancements of the MultiMiner project and highlights of application of the novel machine learning algorithms in selected case studies for mineral exploration and mine site monitoring.In the MultiMiner project, robust, transferable, scalable and automated tools are developed for mineral exploration. These tools are based on multi-source EO data at multiple data scales and platforms and are implemented into a stand-alone software. The tools include a Mineral Mapping Algorithm (MMA) to perform an automatic spectral feature extraction from deposit-type related reference spectra from a customized reference mineral spectral library. Additionally, workflows to perform automated machine learning interpretation of the multiscale EO data mapping results are developed to produce value added mineral maps of alteration zone or proxy minerals. Finally, a Mineral Prospectivity Wizard GUI is developed, facilitating multi-scale mineral mapping and automatic data interpretation in a guided step-by-step process to analyse EO data even usable for non-remote sensing experts.  The developed algorithms are expected to improve accuracy and time-efficiency of direct mineral identification of CRMs and other raw materials.To reduce disruptions to mining operations and monitor environmental aspects of operating and closed mine sites, MultiMiner creates timely mine site monitoring methods. A novel Generic Mine Site Monitoring (GMSM) algorithm, capable of combining multi-source EO data at various temporal, spatial and spectral resolutions, and requiring only a limited amount of in situ data, is developed. The GMSM algorithm leverages EO foundation models for different modalities, and includes support of temporal information as well. The GMSM algorithm can automatically monitor impacts of mining on the environment, such as water quality and acid mine drainage mapping, or combined monitoring of atmospheric and surface dust. Furthermore, success of rehabilitation activities, including monitoring the revegetation status and Tailings Storage Facility (TSF) dismantling are researched. EO-based solutions for improving mining safety and mitigating operational risks are proposed in terms of ground moisture monitoring and open pit and TSF dam stability monitoring.To unlock the potential of EO data, including Copernicus Sentinel-1 and Sentinel-2, EnMAP, drone-borne hyperspectral, radiometric and multiband SAR as well as in situ collected spectral data, we present case studies to demonstrate and validate the use of the MultiMiner machine learning -based algorithms at five test sites in Europe. The acquired field data are harmonized following project-specific guidelines and subsequently, the metadata of the thus acquired field data are safeguarded in a project database. In the presentation, we give a brief overview of the guidelines and the database.  

The management of mineral and coal resources is affected by the shift in the paradigm of regional government administration, which now emphasizes aspects of regional autonomy. Law Number 23, 2014 concerning Regional Government, which genuinely adheres to the maxim of maximal regional autonomy, appears to be incompatible with Law Number 3, 2020 concerning the revision of Law Number 4, 2009 on Mineral and Coal Mining. In practice, this results in a variety of interpretations of the nature and significance of regional autonomy, which ultimately has repercussions for the management of mineral and coal resources. This research aims to investigate the significance and nature of regional autonomy in connection to the management of mineral and coal resources. This is doctrinal legal research and a philosophical approach is applied based on legal principles. The findings are the fundamental principles of managing mineral and coal resources in the future era of regional autonomy focuses on returning to the principle of being a state, which means that the use of mineral and coal resources must be in accordance with the ideals of the state outlined in paragraph 4 of the Preamble of the 1945 Constitution, aspects of environmental harmonization and spatial alignment and the application of the principle of proportionality in regulating government and local government authorities.

We investigated the use of full-range (400–2,500 nm) hyperspectral data obtained by sampling foliar reflectances to discriminate 46 plant species in a tropical wetland in Jamaica. A total of 47 spectral variables, including derivative spectra, spectral vegetation indices, spectral position variables, normalized spectra and spectral absorption features, were used for classifying the 46 species. The Mann–Whitney U-test, paired one-way ANOVA, principal component analysis (PCA), random forest (RF) and a wrapper approach with a support vector machine were used as feature selection methods. Linear discriminant analysis (LDA), an artificial neural network (ANN) and a generalized linear model fitted with elastic net penalties (GLMnet) were then used for species separation. For comparison, the RF classifier (denoted as RFa) was also used to separate the species by using all reflectance spectra and spectral indices, respectively, without applying any feature selection. The RFa classifier was able to achieve 91.8% and 84.8% accuracy with importance-ranked spectral indices and reflectance spectra, respectively. The GLMnet classifier produced the lowest overall accuracies for feature-selected reflectance spectra data (52–77%) when compared with the LDA and ANN methods. However, when feature-selected spectral indices were used, the GLMnet produced overall accuracies ranging from 79 to 88%, which were the highest among the three classifiers that used feature-selected data. A total of 12 species recorded a 100% producer accuracy, but with spectral indices, and an additional 8 species had perfect producer accuracies, regardless of the input features. The results of this study suggest that the GLMnet classifier can be used, particularly on feature-selected spectral indices, to discern vegetation in wetlands. However, it might be more efficient to use RFa without feature-selected variables, especially for spectral indices.

Land use and land cover (LULC) mapping initiatives are essential to support decision making related to the implementation of different policies. There is a need for timely and accurate LULC maps. However, building them is challenging. LULC changes affect natural areas and local biodiversity. When they cause landscape fragmentation, the mapping and monitoring of changes are affected. Due to this situation, improving the efforts for LULC mapping and monitoring in fragmented biomes and ecosystems is crucial, and the adequate separability of classes is a key factor in this process. We believe that combining multidimensional Earth observation (EO) data cubes and spectral vegetation indices (VIs) derived from the red edge, near-infrared, and shortwave infrared bands provided by the Sentinel-2/MultiSpectral Instrument (S2/MSI) mission reduces uncertainties in area estimation, leading toward more automated mappings. Here, we present a low-cost semi-automated classification scheme created to identify croplands, pasturelands, natural grasslands, and shrublands from EO data cubes and the Surface Reflectance to Vegetation Indexes (sr2vgi) tool to automate spectral index calculation, with both produced in the scope of the Brazil Data Cube (BDC) project. We used this combination of data and tools to improve LULC mapping in the Brazilian Cerrado biome during the 2018–2019 crop season. The overall accuracy (OA) of our results is 88%, indicating the potential of the proposed approach to provide timely and accurate LULC mapping from the detection of different vegetation patterns in time series.

This study presents an integrated remote sensing approach for assessing the ecological consequences of the destruction of the Kakhovka Reservoir in Southern Ukraine. The methodology combines spectral vegetation indices, principal component analysis, and Procrustean analysis to evaluate spatial and functional transformations in vegetation cover following a large-scale anthropo genic disaster. The approach was applied to floodplain ecosystems on Khortytsia Island and adjacent areas using satellite imagery from the Sentinel-2 mission for the years 2022 and 2024. A set of twenty-nine spectral indices, sensitive to vegetation density, pigment composition, water conditions, and soil properties, was employed to identify patterns in plant community dynamics and environmental change. Principal component analysis was utilized to identify the dominant axes of spectral variability, while Procrustean rotations facilitated the detection of significant spatial shifts over time. The results demonstrated strong correlations between changes in vegetation patterns and key ecological indicators, including hemeroby, naturalness, species richness, and functional diversity. Two primary ecological trends were identified. The first trend is associated with ecosystem degradation due to anthropogenic pressure, characterized by increasing hemeroby, a decline in naturalness, and reductions in both functional evenness and functional divergence. The second trend reflects the internal reorganization of plant communities under near-natural conditions, where increases in projective cover and species richness occur alongside a decrease in functional richness. Spectral ind ices, such as the normalized difference vegetation index, the normalized difference chlorophyll index, the red-edge vegetation index, the normalized difference tillage index, and the normalized difference water index, have proven particu larly effective in detecting both degradation and successional processes. This study demonstrates that satellite-based spectral indices can serve as reliable proxies for assessing the functional structure and ecological condition of vegetation. The proposed methodology provides an effective tool for spatially explicit and timely environmental monitoring, thereby supporting evidence-based decision-making in post-disaster landscape management, including the question of restoring water bodies or conserving newly formed floodplain ecosystems. This approach has broad applicability for long-term ecological monitoring, restoration planning, and adaptive ma n agement in regions impacted by significant anthropogenic transformations.

Understanding the complex dynamics of heart rate variability (HRV) during pregnancy is crucial for monitoring both maternal well-being and fetal health. In this study, we use the Multifractal Detrended Fluctuations Analysis approach to investigate HRV patterns in pregnant individuals during sleep based on RR interval maxima (MM fluctuations). In addition, we study the type of multifractality within MM fluctuations, that is, if it arises from a broad probability density function or from varying long-range correlations. Furthermore, to provide a comprehensive view of HRV changes during sleep in pregnancy, classical temporal and spectral HRV indices were calculated at quarterly intervals during sleep. Our study population consists of 21 recordings from nonpregnant women, 18 from the first trimester (early-pregnancy) and 18 from the second trimester (middle-pregnancy) of pregnancy. Results. There are statistically significant differences ( -value < 0.05) in mean heart rate, rms heart rate, mean MM fluctuations, and standard deviation of MM fluctuations, particularly in the third and fourth quarter of sleep between pregnant and non-pregnant states. In addition, the early-pregnancy group shows significant differences ( -value < 0.05) in spectral indices during the first and fourth quarter of sleep compared to the non-pregnancy group. Furthermore, the results of our research show striking similarities in the average multifractal structure of MM fluctuations between pregnant and non-pregnant states during normal sleep. These results highlight the influence of different long-range correlations within the MM fluctuations, which could be primarily associated with the emergence of sleep cycles on multifractality during sleep. Finally, we performed a separability analysis between groups using temporal and spectral HRV indices as features per sleep quarter. Employing only three features after Principal Component Analysis (PCA) to the original feature set, achieving complete separability among all groups appears feasible. Using multifractal analysis, our study provides a comprehensive understanding of the complex HRV patterns during pregnancy, which holds promise for maternal and fetal health monitoring. The separability analysis also provides valuable insights into the potential for group differentiation using simple measures such as mean heart rate, rms heart rate, and mean MM fluctuations or in the transformed feature space based on PCA.

In this study, the feasibility of classifying soybean frogeye leaf spot (FLS) is investigated. Leaf images and hyperspectral reflectance data of healthy and FLS diseased soybean leaves were acquired. First, image processing was used to classify FLS to create a reference for subsequent analysis of hyperspectral data. Then, dimensionality reduction methods of hyperspectral data were used to obtain the relevant information pertaining to FLS. Three single methods, namely spectral index (SI), principal component analysis (PCA), and competitive adaptive reweighted sampling (CARS), along with a PCA and SI combined method, were included. PCA was used to select the effective principal components (PCs), and evaluate SIs. Characteristic wavelengths (CWs) were selected using CARS. Finally, the full wavelengths, CWs, effective PCs, SIs, and significant SIs were divided into 14 datasets (DS1–DS14) and used as inputs to build the classification models. Models’ performances were evaluated based on the classification accuracy for both the overall and individual classes. Our results suggest that the FLS comprised of five classes based on the proportion of total leaf surface covered with FLS. In the PCA and SI combination model, 5 PCs and 20 SIs with higher weight coefficient of each PC were extracted. For hyperspectral data, 20 CWs and 26 effective PCs were also selected. Out of the 14 datasets, the model input variables provided by five datasets (DS2, DS3, DS4, DS10, and DS11) were more superior than those of full wavelengths (DS1) both in support vector machine (SVM) and least squares support vector machine (LS-SVM) classifiers. The models developed using these five datasets achieved overall accuracies ranging from 91.8% to 94.5% in SVM, and 94.5% to 97.3% in LS-SVM. In addition, they improved the classification accuracies by 0.9% to 3.6% (SVM) and 0.9% to 3.7% (LS-SVM).

Remote sensing datasets and methods are suitable for mapping and managing the natural resources like minerals, clean water, and energy and also govern their sustainability nowadays. Hyperspectral (HS) imaging has immense potential for rock type classification, mineral mapping, and identification. This work demonstrates the potential of feature extraction techniques and unsupervised machine learning methods for the space-borne hyperspectral remote sensing data in characterizing and identifying mineral and classifying rock type in Banswara, Rajasthan, India. Feature extraction techniques can reveal variations within the data, which can help identify geological areas, reduce noise, and check the dimensionality of the data. Singular value decomposition (SVD)-based principal component analysis (PCA), kernel PCA (KPCA), minimum noise fraction (MNF), and independent component analysis (ICA) were tested for lithological mapping using recently launched DLR Earth Sensing Imaging Spectrometer Hyperspectral (DESIS) and PRecursore IperSpettrale della Missione Applicativa (PRISMA) data in order to map geologically significant areas. Unsupervised machine learning methods, such as Iterative Self-Organizing Data Analysis Technique (ISODATA) and K-means, were also employed. Vertex component analysis (VCA) was utilized to check for similarity and identify various spectral features. Our work demonstrates the advantages of using feature extraction algorithms such as PCA and KPCA over MNF and ICA in geological mapping and interpretability. We recommend K-means as the preferred method for lithological classification of hyperspectral remote sensing data. Our work highlights the potential of advanced feature extraction algorithms for mineral mapping using hyperspectral data, providing different ways to identify minerals and ultimately leading to better mineral resource management.

Earth observation (EO) satellite data is essential to environmental monitoring. At a national and regional level, the open data cubes harness the power of satellite data by providing application programming interfaces and services to the end-users. The volume and the complexity of satellite observations are increasing, demanding novel approaches for data storing, managing, and processing. High-performance computing (HPC) and cloud platforms may improve Big EO data processing performance. However, it is necessary to consider several vital aspects for efficient and flexible EO data processing, such as the interoperability from cloud-HPC and EO data repositories, automatic provisioning and scaling of cloud-HPC resources, cost-effectiveness, support of new EO data formats and open-source packages, or linkage with data cube platforms. The article proposes a scalable EO data processing platform interoperable from cloud-HPC and EO data repositories. The platform enables linking any data repository supporting web coverage service or SpatioTemporal Asset Catalog Application Programming Interfaces (STAC-API), and any cloud or HPC resource supporting scheduling system API for providing access to the cluster backends.

Accurate soil pH and soil organic carbon (SOC) estimations are vital for sustainable agriculture, as pH affects nutrient availability, and SOC is crucial for soil health and fertility. Hyperspectral imaging provides a faster, non-destructive, and economical alternative to standard soil testing. The study utilizes imaging spectroscopic data from the Africa Soil Information Service (AfSIS) and Land Use and Coverage Area Frame Survey (LUCAS-2009) hyperspectral datasets, capturing spatially distributed spectral information. Machine learning (ML) approaches using high-dimensional spectral bands can be computationally expensive, while those using spectral indices are typically limited to multispectral data. This study addresses these challenges by comparing soil pH and SOC prediction using ML models, with both existing spectral indices and individual hyperspectral bands as input features. Results demonstrate that hyperspectral bands outperform existing indices in predictive accuracy, with R values ranging from 0.8 to 0.94 for both soil pH and SOC. To further enhance prediction performance, this study proposes novel spectral indices-soil pH index (SPI) and soil organic carbon index (SOCI)-specifically designed for hyperspectral data using principal component analysis (PCA) and artificial neural networks (ANN). The proposed SPI and SOCI indices address multicollinearity issues and high dimensionality in raw spectral bands, significantly improving predictive accuracy. The SPI and SOCI indices achieve R values of 0.86 for AfSIS soil pH, 0.945 for LUCAS-2009 soil pH, 0.952 for AfSIS SOC, and 0.963 for LUCAS-2009 SOC. These results show that the proposed spectral indices provide a practical solution for precision agriculture, enhancing soil pH and SOC estimations.