The C2S-MS Floods Dataset is a dataset of global flood events with labeled Sentinel-1 & Sentinel-2 pairs. There are 900 sets (1800 total) of near-coincident Sentinel-1 and Sentinel-2 chips (512 x 512 pixels) from 18 global flood events. Each chip contains a water label for both Sentinel-1 and Sentinel-2, as well as a cloud/cloud shadow mask for Sentinel-2. The dataset was constructed by Cloud to Street in collaboration with and funded by the Microsoft Planetary Computer team.

Microsoft Word New Annotation

## Overview

Microsoft Word New Annotation is a dataset for object detection tasks - it contains Wordfile Annotate annotations for 1,101 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

The Rural Access Index (RAI) is a measure of access, developed by the World Bank in 2006. It was adopted as Sustainable Development Goal (SDG) indicator 9.1.1 in 2015, to measure the accessibility of rural populations. It is currently the only indicator for the SDGs that directly measures rural access.The RAI measures the proportion of the rural population that lives within 2 km of an all-season road. An all-season road is one that is motorable all year, but may be temporarily unavailable during inclement weather (Roberts, Shyam, & Rastogi, 2006). This dataset implements and expands on the most recent official methodology put forward by the World Bank, ReCAP's 2019 RAI Supplemental Guidelines. This is, to date, the only publicly available application of this method at a global scale.MethodologyReCAP's methodology provided new insight on what makes a road all-season and how this data should be handled: instead of removing unpaved roads from the network, the ones that are classified as unpaved are to be intersected with topographic and climatic conditions and, whenever there’s an overlap with excess precipitation and slope, a multiplying factor ranging from 0% to 100% is applied to the population that would access to that road. This present dataset developed by SDSN's SDG Transformation Centre proposes that authorities ability to maintain and remediate road conditions also be taken into account.Data sourcesThe indicator relies on four major items of geospatial data: land cover (rural or urban), population distribution, road network extent and the “all-season” status of those roads.Land cover data (urban/rural distinction)Since the indicator measures the acess rural populations, it's necessary to define what is and what isn't rural. This dataset uses the DegUrba Methodology, proposed by the United Nations Expert Group on Statistical Methodology for Delineating Cities and Rural Areas (United Nations Expert Group, 2019). This approach has been developed by the European Commission Global Human Settlement Layer (GHSL-SMOD) project, and is designed to instil some consistency into the definitions based on population density on a 1-km grid, but adjusted for local situations.Population distributionThe source for population distribution data is WorldPop. This uses national census data, projections and other ancillary data from countries to produce aggregated, 100 m2 population data. Road extentTwo widely recognized road datasets are used: the real-time updated crowd-sourced OpenStreetMap (OSM) or the GLOBIO’s 2018 GRIP database, which draws data from official national sources. The reasons for picking the latter are mostly related to its ability to provide information on the surface (pavement) of these roads, to the detriment of the timeliness of the data, which is restrained to the year 2018. Additionally, data from Microsoft Bing's recent Road Detection project is used to ensure completeness. This dataset is completely derived from machine learning methods applied over satellite imagery, and detected 1,165 km of roads missing from OSM.Roads’ all-season statusThe World Bank's original 2006 methodology defines the term all-season as “… a road that is motorable all year round by the prevailing means of rural transport, allowing for occasional interruptions of short duration”. ReCAP's 2019 methodology makes a case for passability equating to the all-season status of a road, along with the assumption that typically the wet season is when roads become impassable, especially so in steep roads that are more exposed to landslides.This dataset follows the ReCAP methodology by creating an passability index. The proposed use of passability factors relies on the following three aspects:• Surface type. Many rural roads in LICs (and even in large high-income countries including the USA and Australia) are unpaved. As mentioned before, unpaved roads deteriorate rapidly and in a different way to paved roads. They are very susceptible to water ingress to the surface, which softens the materials and makes them very vulnerable to the action of traffic. So, when a road surface becomes saturated and is subject to traffic, the deterioration is accelerated. • Climate. Precipitation has a significant effect on the condition of a road, especially on unpaved roads, which predominate in LICs and provide much of the extended connectivity to rural and poor areas. As mentioned above, the rainfall on a road is a significant factor in its deterioration, but the extent depends on the type of rainfall in terms of duration and intensity, and how well the roadside drainage copes with this. While ReCAP suggested the use of general climate zones, we argue that better spatial and temporal resolutions can be acquired through the Copernicus Programme precipitation data, which is made available freely at ~30km pixel size for each month of the year.• Terrain. The gradient and altitude of roads also has an effect on their accessibility. Steep roads become impassable more easily due to the potential for scour during heavy rainfall, and also due to slipperiness as a result of the road surface materials used. Here this is drawn from slope calculated from SRTM Digital Terrain data.• Road maintenance. The ability of local authorities to remediate damaged caused by precipitation and landslides is proposed as a correcting factor to the previous ones. Ideally this would be measured by the % of GDP invested in road construction and maintenance, but this isn't available for all countries. For this reason, GDP per capita is adopted as a proxy instead. The data range is normalized in such a way that a road maxed out in terms of precipitation and slope (accessibility score of 0.25) in a country at the top of the GDP per capita range is brought back at to the higher end of the accessibility score (0.95), while the accessibility score of a road meeting the same passability conditions in a country which GDP per capita is towards the lower end is kept unchanged.Data processingThe roads from the three aforementioned datasets (Bing, GRIP and OSM) are merged together to them is applied a 2km buffer. The populations falling exclusively on unpaved road buffers are multiplied by the resulting passability index, which is defined as the normalized sum of the aforementioned components, ranging from 0.25 to. 0.9, with 0.95 meaning 95% probability that the road is all-season. The index applied to the population data, so, when calculated, the RAI includes the probability that the roads which people are using in each area will be all-season or not. For example, an unpaved road in a flat area with low rainfall would have an accessibility factor of 0.95, as this road is designed to be accessible all year round and the environmental effects on its impassability are minimal.The code for generating this dataset is available on Github at: https://github.com/sdsna/rai

A dataset containing 3,346 synthetically generated RGB images of road segments with cracks. Road segments and crack formations created in Blender, data collected in Microsoft AirSim. Data is split into train (~70%), test (~15%), and validation (~15%) folders. Contains ground truth bounding boxes labelling cracks in both YOLO and COCO JSON format.

This layer displays a global map of land use/land cover (LULC) derived from ESA Sentinel-2 imagery at 10m resolution. Each year is generated with Impact Observatory’s deep learning AI land classification model, trained using billions of human-labeled image pixels from the National Geographic Society. The global maps are produced by applying this model to the Sentinel-2 Level-2A image collection on Microsoft’s Planetary Computer, processing over 400,000 Earth observations per year.The algorithm generates LULC predictions for nine classes, described in detail below. The year 2017 has a land cover class assigned for every pixel, but its class is based upon fewer images than the other years. The years 2018-2024 are based upon a more complete set of imagery. For this reason, the year 2017 may have less accurate land cover class assignments than the years 2018-2024. Key Properties Variable mapped: Land use/land cover in 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024Source Data Coordinate System: Universal Transverse Mercator (UTM) WGS84Service Coordinate System: Web Mercator Auxiliary Sphere WGS84 (EPSG:3857)Extent: GlobalSource imagery: Sentinel-2 L2ACell Size: 10-metersType: ThematicAttribution: Esri, Impact ObservatoryAnalysis: Optimized for analysisClass Definitions: ValueNameDescription1WaterAreas where water was predominantly present throughout the year; may not cover areas with sporadic or ephemeral water; contains little to no sparse vegetation, no rock outcrop nor built up features like docks; examples: rivers, ponds, lakes, oceans, flooded salt plains.2TreesAny significant clustering of tall (~15 feet or higher) dense vegetation, typically with a closed or dense canopy; examples: wooded vegetation, clusters of dense tall vegetation within savannas, plantations, swamp or mangroves (dense/tall vegetation with ephemeral water or canopy too thick to detect water underneath).4Flooded vegetationAreas of any type of vegetation with obvious intermixing of water throughout a majority of the year; seasonally flooded area that is a mix of grass/shrub/trees/bare ground; examples: flooded mangroves, emergent vegetation, rice paddies and other heavily irrigated and inundated agriculture.5CropsHuman planted/plotted cereals, grasses, and crops not at tree height; examples: corn, wheat, soy, fallow plots of structured land.7Built AreaHuman made structures; major road and rail networks; large homogenous impervious surfaces including parking structures, office buildings and residential housing; examples: houses, dense villages / towns / cities, paved roads, asphalt.8Bare groundAreas of rock or soil with very sparse to no vegetation for the entire year; large areas of sand and deserts with no to little vegetation; examples: exposed rock or soil, desert and sand dunes, dry salt flats/pans, dried lake beds, mines.9Snow/IceLarge homogenous areas of permanent snow or ice, typically only in mountain areas or highest latitudes; examples: glaciers, permanent snowpack, snow fields.10CloudsNo land cover information due to persistent cloud cover.11RangelandOpen areas covered in homogenous grasses with little to no taller vegetation; wild cereals and grasses with no obvious human plotting (i.e., not a plotted field); examples: natural meadows and fields with sparse to no tree cover, open savanna with few to no trees, parks/golf courses/lawns, pastures. Mix of small clusters of plants or single plants dispersed on a landscape that shows exposed soil or rock; scrub-filled clearings within dense forests that are clearly not taller than trees; examples: moderate to sparse cover of bushes, shrubs and tufts of grass, savannas with very sparse grasses, trees or other plants.NOTE: Land use focus does not provide the spatial detail of a land cover map. As such, for the built area classification, yards, parks, and groves will appear as built area rather than trees or rangeland classes.Usage Information and Best PracticesProcessing TemplatesThis layer includes a number of preconfigured processing templates (raster function templates) to provide on-the-fly data rendering and class isolation for visualization and analysis. Each processing template includes labels and descriptions to characterize the intended usage. This may include for visualization, for analysis, or for both visualization and analysis. VisualizationThe default rendering on this layer displays all classes.There are a number of on-the-fly renderings/processing templates designed specifically for data visualization.By default, the most recent year is displayed. To discover and isolate specific years for visualization in Map Viewer, try using the Image Collection Explorer. AnalysisIn order to leverage the optimization for analysis, the capability must be enabled by your ArcGIS organization administrator. More information on enabling this feature can be found in the ‘Regional data hosting’ section of this help doc.Optimized for analysis means this layer does not have size constraints for analysis and it is recommended for multisource analysis with other layers optimized for analysis. See this group for a complete list of imagery layers optimized for analysis.Prior to running analysis, users should always provide some form of data selection with either a layer filter (e.g. for a specific date range, cloud cover percent, mission, etc.) or by selecting specific images. To discover and isolate specific images for analysis in Map Viewer, try using the Image Collection Explorer.Zonal Statistics is a common tool used for understanding the composition of a specified area by reporting the total estimates for each of the classes. GeneralIf you are new to Sentinel-2 LULC, the Sentinel-2 Land Cover Explorer provides a good introductory user experience for working with this imagery layer. For more information, see this Quick Start Guide.Global land use/land cover maps provide information on conservation planning, food security, and hydrologic modeling, among other things. This dataset can be used to visualize land use/land cover anywhere on Earth. Classification ProcessThese maps include Version 003 of the global Sentinel-2 land use/land cover data product. It is produced by a deep learning model trained using over five billion hand-labeled Sentinel-2 pixels, sampled from over 20,000 sites distributed across all major biomes of the world.The underlying deep learning model uses 6-bands of Sentinel-2 L2A surface reflectance data: visible blue, green, red, near infrared, and two shortwave infrared bands. To create the final map, the model is run on multiple dates of imagery throughout the year, and the outputs are composited into a final representative map for each year.The input Sentinel-2 L2A data was accessed via Microsoft’s Planetary Computer and scaled using Microsoft Azure Batch. CitationKarra, Kontgis, et al. “Global land use/land cover with Sentinel-2 and deep learning.” IGARSS 2021-2021 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2021.AcknowledgementsTraining data for this project makes use of the National Geographic Society Dynamic World training dataset, produced for the Dynamic World Project by National Geographic Society in partnership with Google and the World Resources Institute.

This layer displays a global map of land use/land cover (LULC) derived from ESA Sentinel-2 imagery at 10m resolution. Each year is generated with Impact Observatory’s deep learning AI land classification model, trained using billions of human-labeled image pixels from the National Geographic Society. The global maps are produced by applying this model to the Sentinel-2 Level-2A image collection on Microsoft’s Planetary Computer, processing over 400,000 Earth observations per year.The algorithm generates LULC predictions for nine classes, described in detail below. The year 2017 has a land cover class assigned for every pixel, but its class is based upon fewer images than the other years. The years 2018-2023 are based upon a more complete set of imagery. For this reason, the year 2017 may have less accurate land cover class assignments than the years 2018-2023.Variable mapped: Land use/land cover in 2017, 2018, 2019, 2020, 2021, 2022, 2023Source Data Coordinate System: Universal Transverse Mercator (UTM) WGS84Service Coordinate System: Web Mercator Auxiliary Sphere WGS84 (EPSG:3857)Extent: GlobalSource imagery: Sentinel-2 L2ACell Size: 10-metersType: ThematicAttribution: Esri, Impact ObservatoryWhat can you do with this layer?Global land use/land cover maps provide information on conservation planning, food security, and hydrologic modeling, among other things. This dataset can be used to visualize land use/land cover anywhere on Earth. This layer can also be used in analyses that require land use/land cover input. For example, the Zonal toolset allows a user to understand the composition of a specified area by reporting the total estimates for each of the classes. NOTE: Land use focus does not provide the spatial detail of a land cover map. As such, for the built area classification, yards, parks, and groves will appear as built area rather than trees or rangeland classes.Class definitionsValueNameDescription1WaterAreas where water was predominantly present throughout the year; may not cover areas with sporadic or ephemeral water; contains little to no sparse vegetation, no rock outcrop nor built up features like docks; examples: rivers, ponds, lakes, oceans, flooded salt plains.2TreesAny significant clustering of tall (~15 feet or higher) dense vegetation, typically with a closed or dense canopy; examples: wooded vegetation, clusters of dense tall vegetation within savannas, plantations, swamp or mangroves (dense/tall vegetation with ephemeral water or canopy too thick to detect water underneath).4Flooded vegetationAreas of any type of vegetation with obvious intermixing of water throughout a majority of the year; seasonally flooded area that is a mix of grass/shrub/trees/bare ground; examples: flooded mangroves, emergent vegetation, rice paddies and other heavily irrigated and inundated agriculture.5CropsHuman planted/plotted cereals, grasses, and crops not at tree height; examples: corn, wheat, soy, fallow plots of structured land.7Built AreaHuman made structures; major road and rail networks; large homogenous impervious surfaces including parking structures, office buildings and residential housing; examples: houses, dense villages / towns / cities, paved roads, asphalt.8Bare groundAreas of rock or soil with very sparse to no vegetation for the entire year; large areas of sand and deserts with no to little vegetation; examples: exposed rock or soil, desert and sand dunes, dry salt flats/pans, dried lake beds, mines.9Snow/IceLarge homogenous areas of permanent snow or ice, typically only in mountain areas or highest latitudes; examples: glaciers, permanent snowpack, snow fields.10CloudsNo land cover information due to persistent cloud cover.11RangelandOpen areas covered in homogenous grasses with little to no taller vegetation; wild cereals and grasses with no obvious human plotting (i.e., not a plotted field); examples: natural meadows and fields with sparse to no tree cover, open savanna with few to no trees, parks/golf courses/lawns, pastures. Mix of small clusters of plants or single plants dispersed on a landscape that shows exposed soil or rock; scrub-filled clearings within dense forests that are clearly not taller than trees; examples: moderate to sparse cover of bushes, shrubs and tufts of grass, savannas with very sparse grasses, trees or other plants.Classification ProcessThese maps include Version 003 of the global Sentinel-2 land use/land cover data product. It is produced by a deep learning model trained using over five billion hand-labeled Sentinel-2 pixels, sampled from over 20,000 sites distributed across all major biomes of the world.The underlying deep learning model uses 6-bands of Sentinel-2 L2A surface reflectance data: visible blue, green, red, near infrared, and two shortwave infrared bands. To create the final map, the model is run on multiple dates of imagery throughout the year, and the outputs are composited into a final representative map for each year.The input Sentinel-2 L2A data was accessed via Microsoft’s Planetary Computer and scaled using Microsoft Azure Batch.CitationKarra, Kontgis, et al. “Global land use/land cover with Sentinel-2 and deep learning.” IGARSS 2021-2021 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2021.AcknowledgementsTraining data for this project makes use of the National Geographic Society Dynamic World training dataset, produced for the Dynamic World Project by National Geographic Society in partnership with Google and the World Resources Institute.

This chipped training dataset is over Sylhet and includes high-resolution imagery (.tif format) and corresponding building footprint vector labels (.geojson format) in 256 x 256 pixel tile/label pairs. This dataset is a ramp Tier 2 dataset, meaning it has NOT been thoroughly reviewed and improved. This dataset was produced for the ramp project and contains 16,217 tiles and 135,375 individual buildings. The satellite imagery resolution is 30 cm and was sourced from Maxar ODP 2022 imagery release for a Bangladesh flood event. Dataset keywords: Peri-urban, Rural, River, Agricultural