Appendix for Master's thesis. Additional graphs for the ones exemplary printed in the thesis

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during four flights aboard a DOE B-200 aircraft over Nevada, Arizona, and New Mexico, U.S., from 2007-10-01 to 2007-10-04. A focus of this data collection was the USDA Jornada Experimental Range (Jornada) in southern New Mexico. To complement the programs of ground measurements, JORNEX (JORNada EXperiment) began in 1995 to collect remotely sensed data from aircraft and satellite platforms to provide spatial and temporal data on physical and biological states of the Jornada rangeland. JORNEX uses remote sensing techniques to study arid rangeland and the responses of vegetation to changing hydrologic fluxes and atmospheric driving forces. This deployment was coordinated by the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 10-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

A data collection campaign using the MODIS/ASTER airborne simulator (MASTER) was conducted in the greater Phoenix metropolitan area in July 2011 to collect visible through mid-infrared multispectral imagery. High resolution (7 m/pixel) land surface temperature products for day and night periods were calculated using the mid-infrared bands of data; surface reflectance, albedo, and Normalized Difference Vegetation Index (NDVI) products were calculated using the visible through shortwave infrared band data for 41 select neighborhoods. While the full MASTER dataset has been processed to at-sensor radiance, it did not include native geolocation data. As georeferencing the entire dataset was not possible with funds available, the processed data described above were extracted for the 41 spatially discrete Phoenix Area Social Survey neighborhoods within the MASTER flight boundary.

This data set provides high-resolution surface reflectance, thermal imagery, burn severity metrics, and LiDAR-derived structural measures of forested areas in the Sierra Nevada Mountains, California, USA, collected before and after the August 2013 Rim and September 2014 King mega forest fires. Pre-fire data were paired with post-fire collections to assess pre- and post-fire landscape characteristics and fire severity. Field estimates of fire severity were collected to compare with derived remote sensing indices. Reflectance measurements for the spectroscopic AVIRIS and MASTER sensors are distributed as multi-band geotiffs for each megafire and acquisition date. Derived operational metric products for each sensor are provided in individual GeoTIFFs. GeoTIFFs produced from LiDAR point data depict first order topographic indices and summary statistics of vertical vegetation structure.

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during two flights aboard a DOE B-200 aircraft over California and New Mexico, U.S., on 2001-05-11 to 2001-05-12. This deployment was coordinated by the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 10-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

This dataset is an educational toy dataset for object detection with remote sensing data. The dataset contains aerial imagery of the city of Stuttgart at 20cm ground sample distance together with bounding box labels indicating cars (single class) present in these images. Labels are present in the COCO format.

The dataset is split into three parts:* a training split (1000 images),* a validation split (190 images) and* a test split (190 images).

Each image features a height and width of 128 pixels each, contains RGB bands and is stored in the png format. Image data were cropped from aerial imagery of the city of Stuttgart taken in 2021, which are available through Stadtmessungsamt Stuttgart as open data under the CC BY 4.0 license: https://opendata.stuttgart.de/dataset/luftbilder-2021

This dataset has been labeled by Yilsey Terea Benavides Miranda, Khem Raj Devkota, David Michael Udoh and Gökhan Yücesan as part of the Master of Photogrammetry and Geoinformatics course "Remote Sensing Studio" in the 2024 summer term at the Stuttgart University of Applied Sciences.

This dataset is about: Master tracks in different resolutions from POLAR 6 flight P6_246_HAMAG_2024_2402090101. Please consult parent dataset @ https://doi.org/10.1594/PANGAEA.967666 for more information.

This dataset is about: Master track from POLAR 6 flight P6_246_HAMAG_2024_2402120201 in 1 sec resolution (zipped, 514 KB).

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected as part of the Hyperspectral Infrared Imager (HyspIRI) mission's preparatory airborne campaign during four flights aboard a DOE B-200 and a NASA ER-2 aircraft over California and New Mexico, U.S., 2008-10-20 to 2008-10-29. A focus of this data collection was the USDA Jornada Experimental Range (Jornada) in southern New Mexico. To complement the programs of ground measurements, JORNEX (JORNada EXperiment) began in 1995 to collect remotely sensed data from aircraft and satellite platforms to provide spatial and temporal data on physical and biological states of the Jornada rangeland. JORNEX uses remote sensing techniques to study arid rangeland and the responses of vegetation to changing hydrologic fluxes and atmospheric driving forces. This deployment was coordinated by NASA's Dryden Flight Research Center (DRFC), renamed Armstrong Flight Research Center in 2014, located in Edwards, California, and the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 30-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during seven flights aboard a DOE B-200 aircraft over Nevada and California, U.S., from 2007-08-30 to 2007-09-02. This data collection focused on mapping earthquake faults in southern California. This deployment was coordinated by the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 10-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during 10 flights aboard a DOE B-200 aircraft over California, Nevada, Arizona, New Mexico, and Texas, U.S., on 1999-05-28 to 1999-06-10. This deployment was coordinated by the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 20-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

@inproceedings{komurcu2024change, title={Change detection in satellite imagery using transformer models and machine learning techniques: a comprehensive captioning dataset}, author={Kürşat K{"o}m{"u}rc{"u} and Linas Petkevi{\v{c}}ius}, booktitle={DAMSS: 15th Conference on Data Analysis Methods for Software Systems, Druskininkai, Lithuania, November 28-30, 2024}, pages={56--57}, year={2024}, publisher={Vilniaus universiteto leidykla} }

https://scholar.google.com/citations?view_op=view_citation&hl=en&user=Pf0mz8UAAAAJ&citation_for_view=Pf0mz8UAAAAJ:9yKSN-GCB0IC

This dataset contains image captions of 4 datasets. captions folder contains caption csv files and other folders contain image pairs. Also, there are augmented images inside these folders.

There are 3 columns in csv files: change: 0 or 1. There is a change or not? caption1: Description of first image caption2 : Description of second image

These captions were created using MiniCPM-V model

Links of original datasets:

CLCD: https://github.com/liumency/CropLand-CD DSIFN: https://github.com/GeoZcx/A-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images/tree/master/dataset LEVIR-CD: https://chenhao.in/LEVIR/ S2Looking: https://github.com/S2Looking/Dataset

This is the dataset presented in the paper The Mountain Habitats Segmentation and Change Detection Dataset accepted for publication in the IEEE Winter Conference on Applications of Computer Vision (WACV), Waikoloa Beach, HI, USA, January 6-9, 2015. The full-sized images and masks along with the accompanying files and results can be downloaded here. The size of the dataset is about 2.1 GB.

The dataset is released under the Creative Commons Attribution-Non Commercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/legalcode).

The dataset documentation is hosted on GitHub at the following address: http://github.com/fjean/mhscd-dataset-doc. Direct download links to the latest revision of the documentation are provided below:

• PDF format: http://github.com/fjean/mhscd-dataset-doc/raw/master/mhscd-dataset-doc.pdf
• Text format: http://github.com/fjean/mhscd-dataset-doc/raw/master/mhscd-dataset-doc.rst

An orthoimage is remotely-sensed image data in which displacement of features in the image caused by terrain relief and sensor orientation have been mathematically removed. Orthoimagery combines the image characteristics of a photograph with the geometric qualities of a map. The Landsat Mosaic orthoimagery database contains Landsat Thematic Mapper imagery for the conterminous United States. The more than 700 Landsat scenes have been resampled to a 1-arc-second (approximately 30-meter) sample interval in a geographic coordinate system using the North American Horizontal Datum of 1983. Three bands have been selected from the eight spectral bands available for each frame. These are bands 4 (near-infrared), 3 (red), and 2 (green), typically displayed as red, green, and blue, respectively. The image is a full-resolution (spectral and spatial), 24-bit color-infrared composite that simulates color infrared film as a "false color composite". NOTE: This EML metadata file does not contain important geospatial data processing information. Before using any NWT LTER geospatial data read the arcgis metadata XML file in either ISO or FGDC compliant format, using ArcGIS software (ArcCatalog > description), or by viewing the .xml file provided with the geospatial dataset.

This dataset includes Level 1B (L1B) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during one flight aboard a DOE B-200 aircraft over Catalina Island, California, U.S., on 2005-10-31. This deployment was coordinated by the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at approximately 7-meter spatial resolution. The L1B file format is HDF-4. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.

Data content: Based on the VIIRS (Visible Infrared Imaging Radiometer Suite) sensor medium resolution 375mNPP-VIIRS active thermal anomaly data, field research, and other big data of the earth, we constructed the global continental region of high-energy-consuming industrial heat source product data set, totaling 25,544 data. After validation 23232 items are industrial heat source objects, and the recognition accuracy is 90.95%. The output format is shapefile.

Time range of data:2012-2021
Spatial scope: Global continental area
Projection method: WGS84
Volume of data: The total volume of data is about 3346kb.
Type of data: Vector

spatial

Summary:

The files contained herein represent green roof footprints in NYC visible in 2016 high-resolution orthoimagery of NYC (described at https://github.com/CityOfNewYork/nyc-geo-metadata/blob/master/Metadata/Metadata_AerialImagery.md). Previously documented green roofs were aggregated in 2016 from multiple data sources including from NYC Department of Parks and Recreation and the NYC Department of Environmental Protection, greenroofs.com, and greenhomenyc.org. Footprints of the green roof surfaces were manually digitized based on the 2016 imagery, and a sample of other roof types were digitized to create a set of training data for classification of the imagery. A Mahalanobis distance classifier was employed in Google Earth Engine, and results were manually corrected, removing non-green roofs that were classified and adjusting shape/outlines of the classified green roofs to remove significant errors based on visual inspection with imagery across multiple time points. Ultimately, these initial data represent an estimate of where green roofs existed as of the imagery used, in 2016.

These data are associated with an existing GitHub Repository, https://github.com/tnc-ny-science/NYC_GreenRoofMapping, and as needed and appropriate pending future work, versioned updates will be released here.

Terms of Use:

The Nature Conservancy and co-authors of this work shall not be held liable for improper or incorrect use of the data described and/or contained herein. Any sale, distribution, loan, or offering for use of these digital data, in whole or in part, is prohibited without the approval of The Nature Conservancy and co-authors. The use of these data to produce other GIS products and services with the intent to sell for a profit is prohibited without the written consent of The Nature Conservancy and co-authors. All parties receiving these data must be informed of these restrictions. Authors of this work shall be acknowledged as data contributors to any reports or other products derived from these data.

Associated Files:

As of this release, the specific files included here are:

GreenRoofData2016_20180917.geojson is in the human-readable, GeoJSON format, in geographic coordinates (Lat/Long, WGS84; EPSG 4263).

GreenRoofData2016_20180917.gpkg is in the GeoPackage format, which is an Open Standard readable by most GIS software including Esri products (tested on ArcMap 10.3.1 and multiple versions of QGIS). This dataset is in the New York State Plan Coordinate System (units in feet) for the Long Island Zone, North American Datum 1983, EPSG 2263.

GreenRoofData2016_20180917_Shapefile.zip is a zipped folder containing a Shapefile and associated files. Please note that some field names were truncated due to limitations of Shapefiles, but columns are in the same order as for other files and in the same order as listed below. This dataset is in the New York State Plan Coordinate System (units in feet) for the Long Island Zone, North American Datum 1983, EPSG 2263.

GreenRoofData2016_20180917.csv is a comma-separated values file (CSV) with coordinates for centroids for the green roofs stored in the table itself. This allows for easily opening the data in a tool like spreadsheet software (e.g., Microsoft Excel) or a text editor.

Column Information for the datasets:

Some, but not all fields were joined to the green roof footprint data based on building footprint and tax lot data; those datasets are embedded as hyperlinks below.

fid - Unique identifier

bin - NYC Building ID Number based on overlap between green roof areas and a building footprint dataset for NYC from August, 2017. (Newer building footprint datasets do not have linkages to the tax lot identifier (bbl), thus this older dataset was used). The most current building footprint dataset should be available at: https://data.cityofnewyork.us/Housing-Development/Building-Footprints/nqwf-w8eh. Associated metadata for fields from that dataset are available at https://github.com/CityOfNewYork/nyc-geo-metadata/blob/master/Metadata/Metadata_BuildingFootprints.md.

bbl - Boro Block and Lot number as a single string. This field is a tax lot identifier for NYC, which can be tied to the Digital Tax Map (http://gis.nyc.gov/taxmap/map.htm) and PLUTO/MapPLUTO (https://www1.nyc.gov/site/planning/data-maps/open-data/dwn-pluto-mappluto.page). Metadata for fields pulled from PLUTO/MapPLUTO can be found in the PLUTO Data Dictionary found on the aforementioned page. All joins to this bbl were based on MapPLUTO version 18v1.

gr_area - Total area of the footprint of the green roof as per this data layer, in square feet, calculated using the projected coordinate system (EPSG 2263).

bldg_area - Total area of the footprint of the associated building, in square feet, calculated using the projected coordinate system (EPSG 2263).

prop_gr - Proportion of the building covered by green roof according to this layer (gr_area/bldg_area).

cnstrct_yr - Year the building was constructed, pulled from the Building Footprint data.

doitt_id - An identifier for the building assigned by the NYC Dept. of Information Technology and Telecommunications, pulled from the Building Footprint Data.

heightroof - Height of the roof of the associated building, pulled from the Building Footprint Data.

feat_code - Code describing the type of building, pulled from the Building Footprint Data.

groundelev - Lowest elevation at the building level, pulled from the Building Footprint Data.

qa - Flag indicating a positive QA/QC check (using multiple types of imagery); all data in this dataset should have 'Good'

notes - Any notes about the green roof taken during visual inspection of imagery; for example, it was noted if the green roof appeared to be missing in newer imagery, or if there were parts of the roof for which it was unclear whether there was green roof area or potted plants.

classified - Flag indicating whether the green roof was detected image classification. (1 for yes, 0 for no)

digitized - Flag indicating whether the green roof was digitized prior to image classification and used as training data. (1 for yes, 0 for no)

newlyadded - Flag indicating whether the green roof was detected solely by visual inspection after the image classification and added. (1 for yes, 0 for no)

original_source - Indication of what the original data source was, whether a specific website, agency such as NYC Dept. of Parks and Recreation (DPR), or NYC Dept. of Environmental Protection (DEP). Multiple sources are separated by a slash.

address - Address based on MapPLUTO, joined to the dataset based on bbl.

borough - Borough abbreviation pulled from MapPLUTO.

ownertype - Owner type field pulled from MapPLUTO.

zonedist1 - Zoning District 1 type pulled from MapPLUTO.

spdist1 - Special District 1 pulled from MapPLUTO.

bbl_fixed - Flag to indicate whether bbl was manually fixed. Since tax lot data may have changed slightly since the release of the building footprint data used in this work, a small percentage of bbl codes had to be manually updated based on overlay between the green roof footprint and the MapPLUTO data, when no join was feasible based on the bbl code from the building footprint data. (1 for yes, 0 for no)

For GreenRoofData2016_20180917.csv there are two additional columns, representing the coordinates of centroids in geographic coordinates (Lat/Long, WGS84; EPSG 4263):

xcoord - Longitude in decimal degrees.

ycoord - Latitude in decimal degrees.

Acknowledgements:

This work was primarily supported through funding from the J.M. Kaplan Fund, awarded to the New York City Program of The Nature Conservancy, with additional support from the New York Community Trust, through New York City Audubon and the Green Roof Researchers Alliance.

Reflected signals of Global Navigation Satellite Systems (GNSS) have been investigated for various applications in remote sensing over the last three decades. The overall research field of GNSS reflectometry includes the retrieval of sea ice parameters as an important application. For this purpose, GNSS reflectometry data have been recorded over the Arctic Ocean with a dedicated receiver setup during the MOSAiC expedition (Multidisciplinary drifting Observatory for the Study of Arctic Climate). The setup was mounted on the German research icebreaker Polarstern (AWI, 2017) that drifted during nine months of the expedition with the Arctic sea ice. The here described data set comprises the expedition’s first leg in autumn 2019. It includes the drift period of the ship from 27th September until 14th December at about 82°N to 87°N in the Siberian Sector of the Arctic. The data set is based on essential contributions of setup & data recording (by GFZ), maintenance & data transfer (by AWI and MOSAiC partners), processing to data level 1 & documentation (by DLR-SO). The level 1 data consist of GNSS signal power estimates of the direct and reflected signal. Data appear in event files (netcdf format) sorted into day folders. Each event includes observations of a satellite on a continuous track, here, in a satellite elevation range from min. 1° to max. 45°. A dedicated GNSS reflectometry receiver, of GORS (GNSS Occultation Reflectometry Scatterometry) type, was used for the measurements. It is equipped with four antenna front-ends. A master channel and two slave channels are assigned to the front-ends. The master channel tracks the GNSS signal on the direct link. The slave channels are dedicated for observations of reflection events: one at left-handed (LH) and another one at right-handed (RH) circular polarization. The respective up-looking master antenna and port-side looking slave antenna (dual-polarization) are set up with a short baseline on the ship’s observation deck, about 22 m above the water level. The given ship-based geometry results in events with rather short excess paths of the reflected signal relative to the direct signal, much less than the range of a code chip (about 300 m for GPS L1 C/A). Interferometric pattern of direct and reflected signal contributions are observed in the channel. A separation step is required in further processing.

This dataset includes Level 1B (L1B) and Level 2 (L2) data products from the MODIS/ASTER Airborne Simulator (MASTER) instrument. The spectral data were collected during three flights aboard a NASA ER-2 and two flights on a DOE B-200 aircraft over California and Nevada U.S. from 2002-08-09 to 2002-08-20. This deployment was coordinated by NASA's Dryden Flight Research Center (DRFC), renamed Armstrong Flight Research Center in 2014, located in Edwards, California, and the U.S. Department of Energy's Remote Sensing Laboratory (RSL) located at Nellis Air Force Base near Las Vegas, Nevada. Data products include L1B georeferenced multispectral imagery of calibrated radiance in 50 bands covering wavelengths of 0.460 to 12.879 micrometers at spatial resolution of 10 to 50 m. Derived L2 data products are emissivity in 5 bands in thermal infrared range (8.58 to 12.13 micrometers) and land surface temperature. The L1B file format is HDF-4, and L2 products are provided in ENVI and KMZ formats. In addition, the dataset includes flight paths, spectral band information, instrument configuration, ancillary notes, and summary information for each flight, and browse images derived from each L1B data file.