A Possible Land Cover EAGLE Approach to Overcome Remote Sensing Limitations in the Alps Based on Sentinel-1 and Sentinel-2: The Case of Aosta Valley (NW Italy)

Remote Sensing, Journal Year: 2022, Volume and Issue: 15(1), P. 178 - 178

Published: Dec. 28, 2022

Land cover (LC) maps are crucial to environmental modeling and define sustainable management planning policies. The development of a land mapping continuous service according the new EAGLE legend criteria has become great interest public sector. In this work, tentative approach map overcoming remote sensing (RS) limitations in mountains newest guidelines was proposed. order reach goal, methodology been developed Aosta Valley, NW Italy, due its higher degree geomorphological complexity. Copernicus Sentinel-1 2 data were adopted, exploiting maximum potentialities limits both, processed Google Earth Engine SNAP. Due SAR geometrical distortions, these used only refine urban water surfaces, while for other classes, composite timeseries filtered regularized stack from Sentinel-2 used. GNSS ground truth with training validation sets. Results showed that K-Nearest-Neighbor Minimum Distance classification permit maximizing accuracy reducing errors. Therefore, mixed hierarchical seems be best solution create LC mountain areas strengthen local concerning mapping.

Language: Английский

---

A Scalable Earth Observation Service to Map Land Cover in Geomorphological Complex Areas beyond the Dynamic World: An Application in Aosta Valley (NW Italy)

Applied Sciences, Journal Year: 2022, Volume and Issue: 13(1), P. 390 - 390

Published: Dec. 28, 2022

Earth Observation services guarantee continuous land cover mapping and are becoming of great interest worldwide. The Google Engine Dynamic World represents a planetary example. This work aims to develop service in geomorphological complex areas the Aosta Valley NW Italy, according newest European EAGLE legend starting year 2020. Sentinel-2 data were processed Engine, particularly summer yearly median composite for each band their standard deviation with multispectral indexes, which used perform k-nearest neighbor classification. To better map some classes, minimum distance classification involving NDVI NDRE filtered regularized stacks computed agronomical classes. Furthermore, SAR Sentinel-1 SLC SNAP urban water surfaces improve optical Additionally, deep learning GIS updated datasets components adopted beginning an aerial orthophoto. GNSS ground truth define training validation sets. In order test effectiveness implemented its methodology, overall accuracy was compared other approaches. A mixed hierarchical approach represented best solution effectively overcome remote sensing limitations. conclusion, this may help implementation local policies concerning surveys both at high spatial temporal resolutions, empowering technological transfer alpine realities.

Language: Английский

---

Land Cover Mapping with Convolutional Neural Networks Using Sentinel-2 Images: Case Study of Rome

Land, Journal Year: 2023, Volume and Issue: 12(4), P. 879 - 879

Published: April 13, 2023

Land cover monitoring is crucial to understand land transformations at a global, regional and local level, the development of innovative methodologies necessary in order define appropriate policies management practices. Deep learning techniques have recently been demonstrated as useful method for mapping through classification remote sensing imagery. This research aims test compare predictive models created using convolutional neural networks (CNNs) VGG16, DenseNet121 ResNet50 on multitemporal single-date Sentinel-2 satellite data. The most promising model was VGG16 both with multi-temporal images, which reach an overall accuracy 71% used produce automatically generated EAGLE-compliant map Rome 2019. methodology part activities ISPRA exploits its main products input support In this sense, it first attempt develop high-update-frequency tool dynamic areas be integrated framework Italian territory.

Language: Английский

---

Effect of the Synergetic Use of Sentinel-1, Sentinel-2, LiDAR and Derived Data in Land Cover Classification of a Semiarid Mediterranean Area Using Machine Learning Algorithms

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(2), P. 312 - 312

Published: Jan. 5, 2023

Land cover classification in semiarid areas is a difficult task that has been tackled using different strategies, such as the use of normalized indices, texture metrics, and combination images from dates or sensors. In this paper we present results an experiment three sensors (Sentinel-1 SAR, Sentinel-2 MSI LiDAR), four indices metrics to classify area. Three machine learning algorithms were used: Random Forest, Support Vector Machines Multilayer Perceptron; Maximum Likelihood was used baseline classifier. The synergetic all these sources resulted significant increase accuracy, Forest being model reaching highest accuracy. However, large amount features (126) advises feature selection reduce figure. After Variance Inflation Factor importance, reduced 62. final overall accuracy obtained 0.91 ± 0.005 (α = 0.05) kappa index 0.898 0.006 0.05). Most observed confusions are easily explicable do not represent difference agronomic terms.

Language: Английский

---

Surface biophysical features fusion in remote sensing for improving land crop/cover classification accuracy

The Science of The Total Environment, Journal Year: 2022, Volume and Issue: 838, P. 156520 - 156520

Published: June 7, 2022

Language: Английский

---

Land Consumption Dynamics and Urban–Rural Continuum Mapping in Italy for SDG 11.3.1 Indicator Assessment

Land, Journal Year: 2023, Volume and Issue: 12(1), P. 155 - 155

Published: Jan. 3, 2023

For the first time in human history, over half of world’s population lives urban areas. This rapid growth makes cities more vulnerable, increasing need to monitor dynamics and its sustainability. The aim this work is examine spatial extent areas, identify urban–rural continuum, understand urbanization processes, Sustainable Development Goal 11. In paper, we apply methodology developed by European Commission-Joint Research Center for classification degree Italian territory, using ISPRA land consumption map ISTAT data. analysis shows that availability detailed updated spatialized data essential calculate SDG indicator 11.3.1, which assesses ratio rate rate. Three new indicators are also proposed describe main trends sprawl, analyzing distribution terms infill settlement dispersion. research good results identifying class boundaries describing urbanized landscape, highlighting demographic obtained lends itself a variety applications, such as monitoring consumption, dynamics, or heat islands, assessing presence state green infrastructures context, driving development policies areas toward sustainable choices focused on regeneration.

Language: Английский

---

Assessing the spatial coherence of forest cover indicators from different data sources: A contribution to sustainable development reporting

Ecological Indicators, Journal Year: 2023, Volume and Issue: 158, P. 111498 - 111498

Published: Dec. 29, 2023

The accuracy of forest area estimates has improved over time as a result field/cadastral surveys, enhanced remote sensing techniques, and the effectiveness algorithms for automatical recognition land cover types. However, statistics seem to be less accurate in disaggregated spatial domains such small administrative units. To evaluate contribution different data sources small-area estimation Europe, we compared seven indicators with coverage resolution. analysis considered multiple information from innovative initiatives, Copernicus Land monitoring scheme, traditional (national) surveys. More specifically, study examined coherence these at municipal scale Italy achieve two objectives: (i) assessing overall precision rates (ii) identifying variations associated technical characteristics each source. A econometric approach was used identify divergence determine providers best suited meet requirements environmental reporting desired scale. results reveal that selected show varying degrees internal coherence, some indices displaying strong correlations others delineating heterogeneous patterns. Our highlights importance choosing right source level provides valuable quantifying reliability key aspects sustainable development.

Language: Английский

---

Mapping built infrastructure in semi-arid systems using data integration and open-source approaches for image classification

Remote Sensing Applications Society and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 101472 - 101472

Published: Jan. 1, 2025

Language: Английский

---

Investigation of the Surface Urban Heat Island (SUHI) by two remote sensing-based approaches in Italian regional capitals

Remote Sensing Applications Society and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 101567 - 101567

Published: April 1, 2025

Language: Английский

---

Afforestation monitoring through automatic analysis of 36-years Landsat Best Available Composites

iForest - Biogeosciences and Forestry, Journal Year: 2022, Volume and Issue: 15(4), P. 220 - 228

Published: July 12, 2022

The study of afforestation is crucial to monitor land transformations and represents a central topic in sustainable development procedures, terms climate change, ecosystem services monitoring, planning policies activities. Although surveying important, the assessment growing forests difficult, since cover has different durations depending on species. In this context, remote sensing can be valid instrument evaluate process. Nevertheless, while vast literature forest disturbance exists, only few studies focus almost none directly exploits data. This aims automatically classify non-forest, afforestation, areas using To purpose, we constructed reference dataset 61 polygons that suffered change from non-forest period 1988-2020. data were with Land Use Inventory Italy through photointerpretation orthophotos (1988-2012, spatial resolution 50 × cm) very high-resolution images (2012-2020, 30 cm). Using Landsat Best Available Pixel composites time-series (1984-2020) calculated 52 temporal predictors: four metrics (median, standard deviation, Pearson’s correlation coefficient R, slope) for 13 bands (the six spectral bands, three Spectral Vegetation Indices, Tasseled Cap Indices). verify possibility distinguishing forest, given differences between them minimal, tested models aiming at classifying following categories: (i) non-forest/afforestation, (ii) afforestation/forest, (iii) non-forest/forest (iv) non-forest/afforestation/forest. Temporal predictors used random which was calibrated search, validated k-fold Cross-Validation Overall Accuracy (OAcv), further out-of-bag independent (OAoob). Results illustrate distinction afforestation/forest reaches largest OAcv (87%), followed by (83%), non-forest/afforestation (75%) non-forest/afforestation/forest (72%). OA values confirm difference photosynthetic activity analysed distinguish them. are currently not exploited our results suggest it may support country-level monitoring reporting.

Language: Английский

---