The Remote Sensing Image Captioning Dataset (RSICD) is a dataset for remote sensing image captioning task. It contains more than ten thousands remote sensing images which are collected from Google Earth, Baidu Map, MapABC and Tianditu. The images are fixed to 224X224 pixels with various resolutions. The total number of remote sensing images is 10921, with five sentences descriptions per image.

Remote Sensing

## Overview

Remote Sensing is a dataset for object detection tasks - it contains Highway annotations for 10,000 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.

This dataset provides a presentation that highlights the role NASA research and researchers played in developing a wide range of significant, quantitative ecological applications of satellite data. The presentation by Dr Diane E. Wickland, former NASA Terrestrial Ecology Program Manager and Lead for NASA Carbon Cycle and Ecosystems Focus Area, provides a top-level overview from her perspective of the development and evolution of the program. Dr Wickland joined NASA in 1985 to manage a newly formed Terrestrial Ecosystems Program. Along with other NASA program managers, she was charged with reorienting the program to be less empirical and have a greater focus on first principles, and to prepare for a next generation of earth-observing satellites. As an ecologist, she thought that focusing on important ecological questions and recruiting practicing ecologists to the program would facilitate such a change in directions. The presentation emphasizes the early years of U.S. satellite remote sensing and covers a few highlights after 2005.

Discover the Remote Sensing Object Segmentation Dataset Perfect for GIS, AI driven environmental studies, and satellite image analysis.

The Satellite Ships Segmentation Dataset is a specialized collection for remote sensing applications, derived from high-resolution satellite imagery with dimensions ranging from 14,722 x 20,949 to 38,133 x 14,604 pixels. This dataset is focused on semantic segmentation, featuring annotations for ships including Automatic Identification System (AIS) information and satellite icon notes, facilitating detailed maritime monitoring and analysis.

The primary goal is to forge a high-quality dataset that streamlines the precise and efficient classification of land use and land cover categories through remote sensing imagery.

The Remote Sensing Satellites Market is segmented by Satellite Mass (10-100kg, 100-500kg, 500-1000kg, Below 10 Kg, above 1000kg), by Orbit Class (GEO, LEO, MEO), by Satellite Subsystem (Propulsion Hardware and Propellant, Satellite Bus & Subsystems, Solar Array & Power Hardware, Structures, Harness & Mechanisms), by End User (Commercial, Military & Government) and by Region (Asia-Pacific, Europe, North America). Market Value in USD is presented. Key Data Points observed include spending on space programs in USD by region; and, the count of satellite launches by satellite launch mass.

Satellite Remote Sensing Image Market Outlook

In 2023, the global satellite remote sensing image market size was valued at approximately USD 3.5 billion and is projected to reach around USD 7.8 billion by 2032, growing at a compound annual growth rate (CAGR) of 9.1% during the forecast period. The market's robust growth can be attributed to an increasing demand for high-resolution satellite imagery, advancements in satellite technology, and a growing number of applications across various industries.

The primary growth factor for the satellite remote sensing image market is the increasing utilization of satellite imagery in diverse fields such as agriculture, environmental monitoring, and disaster management. Farmers and agricultural professionals are increasingly reliant on satellite images to monitor crop health, forecast yields, and manage resources more effectively. The advanced capabilities of remote sensing technology allow for precise monitoring of vegetation, soil moisture, and crop conditions, thereby aiding in precision agriculture. This trend not only helps in maximizing agricultural output but also in making farming more sustainable by optimizing the use of water and fertilizers.

Environmental monitoring is another significant area driving the market growth. Governments and organizations worldwide are leveraging satellite remote sensing to track environmental changes and address climate change issues. By monitoring deforestation, glacier retreat, sea-level rise, and other ecological metrics, stakeholders can make informed decisions to mitigate adverse impacts. The precision and comprehensive coverage offered by satellite imagery make it indispensable for long-term environmental monitoring and conservation efforts.

Disaster management is a critical application where satellite remote sensing images play a pivotal role. Natural disasters such as hurricanes, earthquakes, floods, and wildfires can be swiftly assessed and managed using high-resolution satellite images. These images provide real-time data that enable authorities to allocate resources efficiently, plan evacuations, and assess damage. The ability to quickly and accurately assess disaster impacts helps in reducing response times and improving the effectiveness of relief operations.

The integration of Atmospheric Satellite technology into the field of remote sensing is poised to revolutionize the way we gather data about the Earth's atmosphere. Atmospheric Satellites, often referred to as 'atmosats', are designed to operate at altitudes higher than traditional aircraft but lower than conventional satellites. This unique positioning allows them to capture detailed atmospheric data, which is crucial for understanding weather patterns, climate change, and environmental phenomena. The ability of atmosats to remain in the same position relative to the Earth's surface provides continuous monitoring capabilities, making them invaluable for applications such as real-time weather forecasting, air quality monitoring, and disaster response. As the demand for precise atmospheric data grows, the role of Atmospheric Satellites in enhancing the accuracy and reliability of satellite remote sensing images becomes increasingly significant.

Regionally, North America and Europe are leading the market due to their advanced satellite technologies and high adoption rates across various industries. The presence of key market players, coupled with extensive investments in space programs, propels the market in these regions. North America, in particular, benefits from increased funding for space exploration and defense applications, while Europe focuses on environmental monitoring and urban planning. However, Asia Pacific is expected to witness the highest growth rate during the forecast period, driven by rapid economic development, expanding agricultural activities, and growing investments in space technology.

Technology Analysis

The technology segment of the satellite remote sensing image market is categorized into Optical, Radar, and Hyperspectral imaging technologies. Each of these technologies offers unique advantages and applications, driving their adoption across various sectors. Optical imaging, which involves capturing images using visible light, is widely used for its high-resolution capabilities and cost-effectiveness. This technology is particularly beneficial for applications requiring detailed visual information, such as urban planning, construction monitoring, and

MULTI-TEMPORAL REMOTE SENSING IMAGE CLASSIFICATION - A MULTI-VIEW APPROACH VARUN CHANDOLA AND RANGA RAJU VATSAVAI Abstract. Multispectral remote sensing images have been widely used for automated land use and land cover classification tasks. Often thematic classification is done using single date image, however in many instances a single date image is not informative enough to distinguish between different land cover types. In this paper we show how one can use multiple images, collected at different times of year (for example, during crop growing season), to learn a better classifier. We propose two approaches, an ensemble of classifiers approach and a co-training based approach, and show how both of these methods outperform a straightforward stacked vector approach often used in multi-temporal image classification. Additionally, the co-training based method addresses the challenge of limited labeled training data in supervised classification, as this classification scheme utilizes a large number of unlabeled samples (which comes for free) in conjunction with a small set of labeled training data.

LISTOS_AircraftRemoteSensing_NASAAircraft_Data is the Long Island Sound Tropospheric Ozone Study (LISTOS) remote sensing data collected onboard the NASA aircraft during the LISTOS field campaign. This product is a result of a joint effort across multiple agencies, including NASA, NOAA, the EPA Northeast States for Coordinated Air Use Management (NESCAUM), Maine Department of Environmental Protection, New Jersey Department of Environmental Protection, New York State Department of Environmental Conservation and several research groups at universities. Data collection is complete. The New York City (NYC) metropolitan area (comprised of portions of New Jersey, New York, and Connecticut in and around NYC) is home to over 20 million people, but also millions of people living downwind in neighboring states. This area continues to persistently have challenges meeting past and recently revised federal health-based air quality standards for ground-level ozone, which impacts the health and well-being of residents living in the area. A unique feature of this chronic ozone problem is the pollution transported in a northeast direction out of NYC over Long Island Sound. The relatively cool waters of Long Island Sound confine the pollutants in a shallow and stable marine boundary layer. Afternoon heating over coastal land creates a sea breeze that carries the air pollution inland from the confined marine layer, resulting in high ozone concentrations in Connecticut and, at times, farther east into Rhode Island and Massachusetts. To investigate the evolving nature of ozone formation and transport in the NYC region and downwind, Northeast States for Coordinated Air Use Management (NESCAUM) launched the Long Island Sound Tropospheric Ozone Study (LISTOS). LISTOS was a multi-agency collaborative study focusing on Long Island Sound and the surrounding coastlines that continually suffer from poor air quality exacerbated by land/water circulation. The primary measurement observations took place between June-September 2018 and include in-situ and remote sensing instrumentation that were integrated aboard three aircraft, a network of ground sites, mobile vehicles, boat measurements, and ozonesondes. The goal of LISTOS was to improve the understanding of ozone chemistry and sea breeze transported pollution over Long Island Sound and its coastlines. LISTOS also provided NASA the opportunity to test air quality remote sensing retrievals with the use of its airborne simulators (GEOstationary Coastal and Air Pollution Events (GEO-CAPE) Airborne Simulator (GCAS), and Geostationary Trace gas and Aerosol Sensory Optimization (GeoTASO)) for the preparation of the Tropospheric Emissions; Monitoring of Pollution (TEMPO) observations for monitoring air quality from space. LISTOS also helped collaborators in the validation of Tropospheric Monitoring Instrument (TROPOMI) science products, with use of airborne- and ground-based measurements of ozone, NO2, and HCHO.

This dataset contains imagery and field data associated with the manuscript Garcia-Pineda et al., 2020. It includes high-resolution Synthetic Aperture Radar (SAR) from satellites and the Jet Propulsions Laboratory’s Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) instrument and multispectral satellite imagery that was processed to quantify oil thickness. Original imagery is not available in this dataset because they are proprietary, but the derived products are included. Original data used are listed in the shapefile of oil thickness. Also included in this dataset is the field data collected using an Analytical Spectral Device (ASD) field spectrometer and the oil thickness characteristics from the field. This dataset supports the publication: Garcia-Pineda, O., Staples, G., Jones, C.E., Hu, C., Holt, B., Kourafalou, V., Graettinger, G., DiPinto, L., Ramirez, E., Streett, D., Cho, J., Swayze, G.A., Sun, S., Garcia, D. and Haces-Garcia, F. (2020). Classification of oil spill by thicknesses using multiple remote sensors. Remote Sensing of Environment, 236, 111421. doi:10.1016/j.rse.2019.111421.

Remote sensing based maps of tidal marshes, both of their extents and carbon stocks, have the potential to play a key role in conducting greenhouse gas inventories and implementing climate mitigation policies. Our objective was to generate a single remote sensing model of tidal marsh aboveground biomass and carbon that represents nationally diverse tidal marshes within the conterminous United States (CONUS). To meet this objective we developed the first national-scale dataset of aboveground tidal marsh biomass, species composition, and aboveground plant carbon content (%C) from six CONUS regions: Cape Cod, MA, Chesapeake Bay, MD, Everglades, FL, Mississippi Delta, LA, San Francisco Bay, CA, and Puget Sound, WA. We tested how plant community composition and vegetation structure differences across estuaries influence model development, and whether data from multiple sensors, in particular Sentinel-1 C-band synthetic aperture radar and Landsat, can improve model performance. The final model, driven by six Landsat vegetation indices and with the soil adjusted vegetation index as the most important (n=409, RMSE=464 g/m2, 12.2% normalized RMSE), successfully predicted biomass and carbon for a range of marsh plant functional types defined by height, leaf angle and growth form. Model error was reduced by scaling field measured biomass by Landsat fraction green vegetation derived from object-based classification of National Agriculture Imagery Program imagery. We generated 30m resolution biomass maps for estuarine and palustrine emergent tidal marshes as indicated by a modified NOAA Coastal Change Analysis Program map for each region. With a mean plant %C of 44.1% (n=1384, 95% C.I.=43.99% - 44.37%) we estimated mean aboveground carbon densities (Mg/ha) and total carbon stocks for each wetland type for each region. We applied a multivariate delta method to calculate uncertainties in regional carbon estimates that considered standard error in map area, mean biomass and mean %C. The original version 1.0 of the dataset can be obtained by contacting kbyrd@usgs.gov.

This dataset is part of the larger data collection, “Aerial imagery object identification dataset for building and road detection, and building height estimation”, linked to in the references below and can be accessed here: https://dx.doi.org/10.6084/m9.figshare.c.3290519. For a full description of the data, please see the metadata: https://dx.doi.org/10.6084/m9.figshare.3504413.

Imagery data from the United States Geological Survey (USGS); building and road shapefiles are from OpenStreetMaps (OSM) (these OSM data are made available under the Open Database License: http://opendatacommons.org/licenses/odbl/1.0/); and the Lidar data are from U.S. National Oceanic and Atmospheric Administration (NOAA), the Texas Natural Resources Information System (TNRIS).

Metadata record for data from ASAC Project 291 See the link below for public details on this project.

From the abstracts of the referenced papers:

Ground surveys of the ice sheet in Wilkes Land, Antarctica, have been made on oversnow traverses operating out of Casey. Data collected include surface elevation, accumulation rate, snow temperature, and physical surveys, the data are mostly restricted to line profiles. In some regions, aerial surveys of surface topology have been made over a grid network. Satellite imagery and remote sensing are two means of extrapolating the results from measurements along lines to an areal presentation. They are also the only source of data over large areas of the continent. Landsat images in the visible and near infra-red wavelengths clearly depict many of the large- and small-scale features of the surface. The intensity of the reflected radiation varies with the aspect and magnitude of the surface slope to reveal the surface topography. The multi-channel nature of the Landsat data are exploited to distinguish between different surface types through their different spectral signatures, e.g. bare ice, glaze, snow, etc. Additional information on surface type can be gained at a coarser scale from other satellite-borne sensors such as the ESMR, SMMR, etc. Textural enhancement of the Landsat images reveals the surface micro-relief. Features in the enhanced images are compared to ground-truth data from the traverse surveys to produce a classification of the surface types across the images and to determine the magnitude of the surface topography and micro-relief observed. The images can then be used to monitor changes over time.

Landsat imagery of the Antarctic ice sheet and glaciers exhibit features that move with the ice and others that are fixed in space. Two images covering the same area but acquired at different times are compared to obtain the displacement of features. Where the time lapse is large, the displacement of obvious features can be scaled from photographic prints. When the two images are co-registered finer features and displacements can be resolved to give greater detail.

Remote sensing techniques can be used to investigate the dynamics and surface characteristics of the Antarctic ice sheet and its outlet glaciers. This paper describes a methodology developed to map glacial movement velocities from LANDSAT MSS data, together with an assessment of the accuracy achieved. The velocities are derived by using digital image processing to register two temporally separated LANDSAT images of the Denman glacier and Shackleton Ice Shelf region. A derived image map is compared with existing maps of the region to substantiate the measured velocities. The velocity estimates from this study were found to correspond closely with ground-based measurements in the study area.

Lesson 1. An Introduction to working with multispectral satellite data in ArcGIS Pro In which we learn: • How to unpack tar and gz files from USGS EROS • The basic map interface in ArcGIS • How to add image files • What each individual band of Landsat spectral data looks like • The difference between: o Analysis-ready data: surface reflectance and surface temperature o Landsat Collection 1 Level 3 data: burned area and dynamic surface water o Sentinel2data o ISRO AWiFS and LISS-3 data Lesson 2. Basic image preprocessing In which we learn: • How to composite using the composite band tool • How to represent composite images • All about band combinations • How to composite using raster functions • How to subset data into a rectangle • How to clip to a polygon Lesson 3. Working with mosaic datasets In which we learn: o How to prepare an empty mosaic dataset o How to add images to a mosaic dataset o How to change symbology in a mosaic dataset o How to add a time attribute o How to add a time dimension to the mosaic dataset o How to view time series data in a mosaic dataset Lesson 4. Working with and creating derived datasets In which we learn: • How to visualize Landsat ARD surface temperature • How to calculate F° from K° using ARD surface temperature • How to generate and apply .lyrx files • How to calculate an NDVI raster using ISRO LISS-3 data • How to visualize burned areas using Landsat Level 3 data • How to visualize dynamic surface water extent using Landsat Level 3 data

The scarcity of specialized datasets and annotation tools presents a formidable hurdle in advancing high-resolution satellite image cloud segmentation algorithms, underscoring the urgency to explore innovative annotation strategies. Ground truth label data stands as a linchpin in the realm of deep learning frameworks, facilitating the iterative refinement of model parameters to achieve optimal predictive accuracy. Despite the dearth of dedicated datasets and annotation software tailored for high-resolution satellite image cloud segmentation tasks, this study introduces CloudLabel, a semi-automatic annotation technique. Harnessing the methods of region growing and morphological processing, CloudLabel streamlines the annotation process for high-resolution satellite images, thereby addressing the limitations of existing annotation platforms which are not specifically adapted to cloud segmentation. By providing a more adaptable and efficient annotation approach compared to conventional labeling methods, CloudLabel enables precise discrimination between cloud and non-cloud regions in high-resolution satellite imagery. Notably, we have curated a dataset comprising 32,065 images (512*512) for cloud segmentation based on CloudLabel, which significantly contributes to algorithmic advancement and broader applications in remote sensing. The 'data' folder contains 32065 RGB images, while the 'label' contains the corresponding CloudLabel labeling results.

Remote Sensing Deforestation Ds

## Overview

Remote Sensing Deforestation Ds is a dataset for instance segmentation tasks - it contains Tree annotations for 961 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 size and share of the market is categorized based on Type (0.3m Resolution, 0.5m Resolution, Other) and Application (Transportation, Agriculture, Surveying and Exploration, Military and Defense, Other) and geographical regions (North America, Europe, Asia-Pacific, South America, and Middle-East and Africa).

The Remote Sensing Division is responsible for providing data to support the Coastal Mapping Program, Emergency Response efforts, and the Aeronautical Survey Program through the use of remotely sensed data. NOAA Coastal Mapping Remote Sensing Data includes metric-quality aerial photographs from film and digital cameras, orthomosaics, and Light Detection and Ranging (lidar). The predecessors to...