{"_id":"5ac295e13934f4002f1a5a62","project":"55faeacad0e22017005b8265","version":{"_id":"55faeacad0e22017005b8268","project":"55faeacad0e22017005b8265","__v":36,"createdAt":"2015-09-17T16:31:06.800Z","releaseDate":"2015-09-17T16:31:06.800Z","categories":["55faeacbd0e22017005b8269","55faf550764f50210095078e","55faf5b5626c341700fd9e96","55faf8a7825d5f19001fa386","560052f91503430d007cc88f","560054f73aa0520d00da0b1a","56005aaf6932a00d00ba7c62","56005c273aa0520d00da0b3f","5601ae7681a9670d006d164d","5601ae926811d00d00ceb487","5601aeb064866b1900f4768d","5601aee850ee460d0002224c","5601afa02499c119000faf19","5601afd381a9670d006d1652","561d4c78281aec0d00eb27b6","561d588d8ca8b90d00210219","563a5f934cc3621900ac278c","5665c5763889610d0008a29e","566710a36819320d000c2e93","56ddf6df8a5ae10e008e3926","56e1c96b2506700e00de6e83","56e1ccc4e416450e00b9e48c","56e1ccdfe63f910e00e59870","56e1cd10bc46be0e002af26a","56e1cd21e416450e00b9e48e","56e3139a51857d0e008e77be","573b4f62ef164e2900a2b881","57c9d1335fd8ca0e006308ed","57e2bd9d1e7b7220000d7fa5","57f2b992ac30911900c7c2b6","58adb5c275df0f1b001ed59b","58c81b5c6dc7140f003c3c46","595412446ed4d9001b3e7b37","59e76ce41938310028037295","5a009de510890d001c2aabfe","5a96f89c89442e002041144b"],"is_deprecated":false,"is_hidden":false,"is_beta":false,"is_stable":true,"codename":"v1","version_clean":"1.0.0","version":"1"},"category":{"_id":"5ac2956b2e75480053d84c36","project":"55faeacad0e22017005b8265","version":"55faeacad0e22017005b8268","__v":0,"sync":{"url":"","isSync":false},"reference":false,"createdAt":"2018-04-02T20:41:15.403Z","from_sync":false,"order":4,"slug":"raster-data-access-guide","title":"Raster Data Access (RDA) Guide"},"user":"55fae9d4825d5f19001fa379","__v":0,"parentDoc":null,"updates":[],"next":{"pages":[],"description":""},"createdAt":"2018-04-02T20:43:13.328Z","link_external":false,"link_url":"","githubsync":"","sync_unique":"","hidden":false,"api":{"results":{"codes":[]},"settings":"","auth":"required","params":[],"url":""},"isReference":false,"order":0,"body":"# Table of Contents\n\nSection | Description\n--- | ---\n[Overview](#section-overview) | Learn more about the powerful and easy-to-use raster data access framework, designed for GBDX users to build and run analytics that work at any scale.\n[Quickstart Tutorial](#section-quickstart-tutorial-for-getting-an-rda-chip-using-gbdxtools) | This quickstart tutorial illustrates how to get an RDA chip using gbdxtools.\n[Additional Resources](#section-additional-resources) | Find links to more RDA documentation.\n\n#Overview\nGBDX Raster Data Access (RDA) is a powerful yet easy-to-use raster data access framework designed for GBDX users to build and run analytics that work at any scale. RDA provides scalable high-performance access to raster data for any area-of-interest (AOI) with dynamic on-the-fly processing. The processing capabilities of RDA include all of the undifferentiated geospatial heavy lifting imagery preprocessing corrections and enhancements to make the raster data suitable for both on-screen visualization and automated analysis. By using GBDX Raster Data Access you get access to the exact imagery data you need when you need it.\n \nRaster Data Access supports imagery data from DigitalGlobe's industry-leading satellite constellation including QuickBird, GeoEye-1, and all WorldView platforms along with other raster data content available on the GBDX platform. RDA provides random AOI-based access to all the raster pixel data, image properties and sensor metadata needed to develop automated analysis algorithms. Furthermore, RDA puts the power in the hands of the developer to apply the exact imagery preprocessing they need for their algorithm. RDA also provides artificial intelligence (AI) developers with quick-and-easy access to image chips that can be used to generate training data for machine learning algorithms or more complex deep learning models.\n \nRaster Data Access was built from the ground-up as microservices using a cloud-native architecture that runs efficiently on AWS. RDA leverages the IDAHO cloud storage format for imagery and metadata which enables distributed computing and parallel processing. The combination of RDA's powerful cloud-computing together with the massive library of geospatial big data available on the GBDX platform allows users to derive answers at a scale that has been technically and economically impossible to achieve until now. RDA enables cloud-based access to DigitalGlobe's image library along with automated analytics which extract location intelligence and actionable insights at a global scale.\n \nRDA can be utilized through a variety of simple APIs and client interfaces including Python (gbdxtools), GBDX Notebooks, and the GDAL software library. With RDA, GBDX users have on-demand control of a wide variety of imagery preprocessing corrections and enhancements,  including but not limited to:\n\n* Orthorectification (any coordinate system via EPSG code)\n* TOA At-Sensor Reflectance (top-of-atmosphere)\n* AComp (atmospheric compensation)*\n* Pan-Sharpening (locally projective algorithm)\n* Dynamic Range Adjustment (radiometric stretch)\n \n*AComp is a fully automated framework for atmospherically compensating very high spatial resolution imagery. AComp converts imagery to surface reflectance units that enables the extraction of information using physical quantities and improves multi-temporal or cross-sensor data analysis. For more detailed background information on DigitalGlobe's proprietary AComp methodology please refer to the following resource:\n\nhttps://calval.cr.usgs.gov/wordpress/wp-content/uploads/AComp_validation.pdf\n\n[block:image]\n{\n  \"images\": [\n    {\n      \"image\": [\n        \"https://files.readme.io/3db80aa-AComp_image_jpeg.jpg\",\n        \"AComp image jpeg.jpg\",\n        1021,\n        389,\n        \"#6a6b62\"\n      ],\n      \"caption\": \"This illustration shows an image before and after atmospheric compensation (AComp)\"\n    }\n  ]\n}\n[/block]\n#Quickstart  Tutorial for Getting an RDA Chip using gbdxtools\n\nGbdxtools is a python client library for accessing GBDX services, including RDA.  Gbdxtools makes it easy for python developers to get basic image chips.\n\nMore information about gbdxtools, see [gbdxtools installation and troubleshooting ](https://github.com/DigitalGlobe/gbdxtools#gbdxtools-python-tools-for-using-gbdx) and [gbdxtools user documentation](http://gbdxtools.readthedocs.io/en/latest/).\n\nThe CatalogImage class is the most common starting point.  For this tutorial, we'll be access the image '10400E0001DB6A00'.  It looks like this:\n[block:image]\n{\n  \"images\": [\n    {\n      \"image\": [\n        \"https://files.readme.io/feeb132-RDA_full_strip_image_1.png\",\n        \"RDA_full_strip_image_1.png\",\n        1752,\n        1456,\n        \"#9f9694\"\n      ],\n      \"caption\": \"Image 1: This is the original image used in the example.\"\n    }\n  ]\n}\n[/block]\n\nIn this tutorial we will get a chip from this image.  The footprint of the chip is shown here in blue:\n[block:image]\n{\n  \"images\": [\n    {\n      \"image\": [\n        \"https://files.readme.io/0a81a5e-RDA-prjwin_footprint-image_2.png\",\n        \"RDA-prjwin_footprint-image 2.png\",\n        1742,\n        1448,\n        \"#9f9694\"\n      ],\n      \"caption\": \"Image 2: The footprint of the chip is shown here in blue.\"\n    }\n  ]\n}\n[/block]\nThe final image chip should look like this:\n\n[block:image]\n{\n  \"images\": [\n    {\n      \"image\": [\n        \"https://files.readme.io/a0e203e-RDA-prjwin_example-image_3.png\",\n        \"RDA-prjwin_example-image 3.png\",\n        1470,\n        1454,\n        \"#c3c6cd\"\n      ],\n      \"caption\": \"Image 3: This is the final image chip produced from the gbdxtools tutorial.\"\n    }\n  ]\n}\n[/block]\nGet image [metadata](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#gbdxtools.catalog.Catalog.get_strip_metadata):\n\n```python\nfrom gbdxtools.catalog import Catalog\nfrom gbdxtools.images.catalog_image import CatalogImage\n\nimageId = \"10400E0001DB6A00\"\ncrop = \"POLYGON ((36.9392906658426 33.9164103054402,36.943294798074 33.9164103054402,36.943294798074 33.9124061732088,36.9392906658426 33.9124061732088,36.9392906658426 33.9164103054402))\"\n\nprint(Catalog().get_strip_metadata(imageId))\n```\n\n[Get](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#geotiff) and [display](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#plot) image chip:\n\n```python\nc = CatalogImage(imageId, product=\"ortho\")\naoi = c.aoi(wkt=crop)\nimage = aoi.geotiff(path=\"output.tif\")\naoi.plot()\n```\n\nBy supplying arguments to the CatalogImage constructor, you can control what type of processing is applied to the image chips.  For details on the arguments available, please visit the CatalogImage [docs](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#catalogimage).\n\n#Additional Resources\n\n## GBDX Notebooks \n\nGBDX Notebooks lets you code python to access imagery and run analytics in a Jupyter-hosted environment. Imagery access in GBDX Notebooks is powered by RDA. \n\nTo learn more about accessing imagery using GBDX Notebooks, see the [Ordering and Working with Imagery](https://notebooks.geobigdata.io/hub/tutorials/5a0370dfe21cea77cee2436b?tab=code) tutorial.\n\n## GDAL 2.3.0 with RDA driver\nThe GDAL 2.3.0 release will include the RDA driver. GDAL 2.3.0 is currently a pre-release version, and is not fully supported. \n\nTo try out the RDA driver with the pre-release 2.3.0 version, see the following tutorials:\n\n[How to Get the GDAL RDA Driver](doc:how-to-get-the-rda-gdal-driver) \n[How to Get a Chip with GDAL and the RDA Template API](doc:how-to-get-a-chip-with-gdal) \n\n## gbdxtools\nTo learn more about the image classes supported by gbdxtools, see [Image Classes](http://gbdxtools.readthedocs.io/en/latest/image_classes.html).","excerpt":"This course provides an overview of Raster Data Access (RDA), and describes how to use it. A quickstart script for getting an RDA chip with gbdxtools is included.","slug":"raster-data-access-course","type":"basic","title":"Raster Data Access (RDA) Course"}

Raster Data Access (RDA) Course

This course provides an overview of Raster Data Access (RDA), and describes how to use it. A quickstart script for getting an RDA chip with gbdxtools is included.

# Table of Contents Section | Description --- | --- [Overview](#section-overview) | Learn more about the powerful and easy-to-use raster data access framework, designed for GBDX users to build and run analytics that work at any scale. [Quickstart Tutorial](#section-quickstart-tutorial-for-getting-an-rda-chip-using-gbdxtools) | This quickstart tutorial illustrates how to get an RDA chip using gbdxtools. [Additional Resources](#section-additional-resources) | Find links to more RDA documentation. #Overview GBDX Raster Data Access (RDA) is a powerful yet easy-to-use raster data access framework designed for GBDX users to build and run analytics that work at any scale. RDA provides scalable high-performance access to raster data for any area-of-interest (AOI) with dynamic on-the-fly processing. The processing capabilities of RDA include all of the undifferentiated geospatial heavy lifting imagery preprocessing corrections and enhancements to make the raster data suitable for both on-screen visualization and automated analysis. By using GBDX Raster Data Access you get access to the exact imagery data you need when you need it. Raster Data Access supports imagery data from DigitalGlobe's industry-leading satellite constellation including QuickBird, GeoEye-1, and all WorldView platforms along with other raster data content available on the GBDX platform. RDA provides random AOI-based access to all the raster pixel data, image properties and sensor metadata needed to develop automated analysis algorithms. Furthermore, RDA puts the power in the hands of the developer to apply the exact imagery preprocessing they need for their algorithm. RDA also provides artificial intelligence (AI) developers with quick-and-easy access to image chips that can be used to generate training data for machine learning algorithms or more complex deep learning models. Raster Data Access was built from the ground-up as microservices using a cloud-native architecture that runs efficiently on AWS. RDA leverages the IDAHO cloud storage format for imagery and metadata which enables distributed computing and parallel processing. The combination of RDA's powerful cloud-computing together with the massive library of geospatial big data available on the GBDX platform allows users to derive answers at a scale that has been technically and economically impossible to achieve until now. RDA enables cloud-based access to DigitalGlobe's image library along with automated analytics which extract location intelligence and actionable insights at a global scale. RDA can be utilized through a variety of simple APIs and client interfaces including Python (gbdxtools), GBDX Notebooks, and the GDAL software library. With RDA, GBDX users have on-demand control of a wide variety of imagery preprocessing corrections and enhancements, including but not limited to: * Orthorectification (any coordinate system via EPSG code) * TOA At-Sensor Reflectance (top-of-atmosphere) * AComp (atmospheric compensation)* * Pan-Sharpening (locally projective algorithm) * Dynamic Range Adjustment (radiometric stretch) *AComp is a fully automated framework for atmospherically compensating very high spatial resolution imagery. AComp converts imagery to surface reflectance units that enables the extraction of information using physical quantities and improves multi-temporal or cross-sensor data analysis. For more detailed background information on DigitalGlobe's proprietary AComp methodology please refer to the following resource: https://calval.cr.usgs.gov/wordpress/wp-content/uploads/AComp_validation.pdf [block:image] { "images": [ { "image": [ "https://files.readme.io/3db80aa-AComp_image_jpeg.jpg", "AComp image jpeg.jpg", 1021, 389, "#6a6b62" ], "caption": "This illustration shows an image before and after atmospheric compensation (AComp)" } ] } [/block] #Quickstart Tutorial for Getting an RDA Chip using gbdxtools Gbdxtools is a python client library for accessing GBDX services, including RDA. Gbdxtools makes it easy for python developers to get basic image chips. More information about gbdxtools, see [gbdxtools installation and troubleshooting ](https://github.com/DigitalGlobe/gbdxtools#gbdxtools-python-tools-for-using-gbdx) and [gbdxtools user documentation](http://gbdxtools.readthedocs.io/en/latest/). The CatalogImage class is the most common starting point. For this tutorial, we'll be access the image '10400E0001DB6A00'. It looks like this: [block:image] { "images": [ { "image": [ "https://files.readme.io/feeb132-RDA_full_strip_image_1.png", "RDA_full_strip_image_1.png", 1752, 1456, "#9f9694" ], "caption": "Image 1: This is the original image used in the example." } ] } [/block] In this tutorial we will get a chip from this image. The footprint of the chip is shown here in blue: [block:image] { "images": [ { "image": [ "https://files.readme.io/0a81a5e-RDA-prjwin_footprint-image_2.png", "RDA-prjwin_footprint-image 2.png", 1742, 1448, "#9f9694" ], "caption": "Image 2: The footprint of the chip is shown here in blue." } ] } [/block] The final image chip should look like this: [block:image] { "images": [ { "image": [ "https://files.readme.io/a0e203e-RDA-prjwin_example-image_3.png", "RDA-prjwin_example-image 3.png", 1470, 1454, "#c3c6cd" ], "caption": "Image 3: This is the final image chip produced from the gbdxtools tutorial." } ] } [/block] Get image [metadata](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#gbdxtools.catalog.Catalog.get_strip_metadata): ```python from gbdxtools.catalog import Catalog from gbdxtools.images.catalog_image import CatalogImage imageId = "10400E0001DB6A00" crop = "POLYGON ((36.9392906658426 33.9164103054402,36.943294798074 33.9164103054402,36.943294798074 33.9124061732088,36.9392906658426 33.9124061732088,36.9392906658426 33.9164103054402))" print(Catalog().get_strip_metadata(imageId)) ``` [Get](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#geotiff) and [display](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#plot) image chip: ```python c = CatalogImage(imageId, product="ortho") aoi = c.aoi(wkt=crop) image = aoi.geotiff(path="output.tif") aoi.plot() ``` By supplying arguments to the CatalogImage constructor, you can control what type of processing is applied to the image chips. For details on the arguments available, please visit the CatalogImage [docs](http://gbdxtools.readthedocs.io/en/latest/api_reference.html?highlight=metadata#catalogimage). #Additional Resources ## GBDX Notebooks GBDX Notebooks lets you code python to access imagery and run analytics in a Jupyter-hosted environment. Imagery access in GBDX Notebooks is powered by RDA. To learn more about accessing imagery using GBDX Notebooks, see the [Ordering and Working with Imagery](https://notebooks.geobigdata.io/hub/tutorials/5a0370dfe21cea77cee2436b?tab=code) tutorial. ## GDAL 2.3.0 with RDA driver The GDAL 2.3.0 release will include the RDA driver. GDAL 2.3.0 is currently a pre-release version, and is not fully supported. To try out the RDA driver with the pre-release 2.3.0 version, see the following tutorials: [How to Get the GDAL RDA Driver](doc:how-to-get-the-rda-gdal-driver) [How to Get a Chip with GDAL and the RDA Template API](doc:how-to-get-a-chip-with-gdal) ## gbdxtools To learn more about the image classes supported by gbdxtools, see [Image Classes](http://gbdxtools.readthedocs.io/en/latest/image_classes.html).