All Tutorial Notebooks

Exploring training data (Embassies) based on OSM geometries in VectorServices and Imagery in PDAP via RADD
Notebook by: Mike Gleason (michael.gleason@digitalglobe.com) In this Notebook, we demonstrate some basic functionality and workflows that can be used for small scale labeling of satellite imagery. These methods can be used to build up small sets of training or test data to help in algorithm development.
In this tutorial by Sean Gorman we'll cover how to find SWIR imagery and use it for creating a Normalized Difference Built-Up Index for analysis of Central Park in Manhattan. This work includes discussion of the difference between VNIR and SWIR data as well as using downsampling techniques to enable the use of band math on imagery products with different resolutions.
This is the second in the series and shows you how to search and bring images into your notebook Written by Steven Poutsy (steven.poutsy@digitalglobe.com)
Notebook by Mike Gleason (michael.gleason@digitalglobe.com) Ecopia Building Footprints powered by Digitalglobe are an off-the-shelf Information Product that are produced by one of DigitalGlobe's Information Partners, Ecopia. In this GBDX Notebook, we explore how products such as GBDX Building Footprints can be combined with the analytical capabilities of GBDX to produce additional data and insights.
Animating imagery over time from a list of CatalogIDs
Methods described here using image processing color slicing method to create binary mask of ships. Ships are then vectorized.
This notebook is the third in a series of beginner tutorial to introduce news users to Jupyter Notebooks, working with satellite imagery and performing basic analytics. Specifically this notebook introduces the concepts of multispectral bands found in satellite imagery and how they can be used for analytics using "band math".
Covers different options for pre-processing on images before they are brought into your notebook. Written by Steve Poutsy (steven.poutsy@digitalglobe.com)
Notebook by: Chloé Hampton (chloe.hamption@digitalglobe.com) and Martha Morrissey (martha.morrissey@digitalglobe.com) A notebook that shows refugee tent growth at the Rukban camp in Syria. Created by Chloé Hampton and Martha Morrissey, with help from Mike Foster, Mike Gleason, and Sean Gorman.
This notebook is a detailed introduction to ways to access DG imagery via the gbdxtools python package.
Demonstrates extracting vectors from classified images Written by Steve Poutsy (steven.poutsy@digitalglobe.com)
This notebook uses a combination of Ecopia Building Footprints and application of the Spectral Angle Mapper (SAM) algorithm to 8-band imagery to estimate impacts from the Carr wildfires. Written by Mike Foster (mike.foster@digitalglobe.com)
A description has not been provided yet
A description has not been provided yet
This notebook builds on color matching concepts discussed in Color Adjustments for RDA Imagery Part 1. Written by Marc Pfister (marc.pfister@digitalglobe.com)
A description has not been provided yet
A notebook by Pramukta Kumar that shows how to histogram match RDA imagery that is native to GBDX Notebooks to DigitalGlobe's Maps API using rio_hist. This method will help provide consistent color balanced imagery that is aesthetically pleasing for viewing.
An introduction to various image classes supported by gbdxtools
Notebook by: Mike Gleason (michael.gleason@digitalglobe.com)
Notebook by: Mike Gleason (michael.gleason@digitalglobe.com) It's easy to think about GBDX and GBDX Notebooks as simply a place to access imagery and do analysis; however, the true power of these tools lies in their potential to produce new information and provide insight to specific research questions or business use-cases. This Notebook demonstrates a simple example of how GBDX can enable you to quickly climb the ladder from imagery to insights. To do so, we start by doing some analysis to extract boats from an image, extend that analysis over the historical archive of imagery in the selected location, and then bring in ancillary data to gain insights into the observed variability in boat traffic.
This notebook tutorial is intended to cover a variety of coding themes to empower new GBDX Notebook developers. Written by Mike Foster (mike.foster@digitalglobe.com)
A description has not been provided yet
This Notebook demonstrates a workflow to detect the installation of temporary blue roofing following Hurricane Maria in Puerto Rico. We then apply that workflow over neighborhoods in multiple towns, and review and interpret the results. Written by Mike Gleason (michael.gleason@digitalglobe.com)
This notebooks by Sean Gorman provides a set of images and data visualization patterns to look at pre and post event imagery for the eruption of the Volcan de Fuergo
This notebook provides a basic introduction to using Dask with satellite imagery. It covers a brief overview of Dask arrays, chunks and map_blocks to scale out an NDVI analysis for a entire strip of imagery over Sydney, Australia.
This notebooks by Sean Gorman provides a search function to look across multiple satellite sensors to find imagery for wildfire analysis. It then shows an analysis pattern for calculating NBR based burn areas and turning them into vector geometries. The methods are portable geographically and should be pluggable into any wildfire scenario to map burn areas.
Notebook by: Laura Atkinson (laura.atkinson@digitalglobe.com) The goal of this notebook is the utilize both DigitalGlobe imagery and LiDAR data to automatically extract trees. LiDAR allows us to create highly detailed 3D models of the earths surface. Areas high in vegetation have high values in the normalized difference vegetation index (NDVI). Also trees are taller than other types of vegetation such as grass and shrubs. By combining these two data sources we can find tall vegetated areas, aka Trees.
Notebook by: Chris Helm (chris.helm@digitalglobe.com) An example of comparing changes in imagery over time in Dubai. This notebook explores ways to collect several images, compute water indices from each, extract vector features, and compare changes to land area over time.
Notebook by Mike Gleason (michael.gleason@digitalglobe.com) This Notebook demonstrates a simple workflow for identifying standing water in an image, extracting and rasterizing roads vectors from OpenStreetMap, and assessing the extent of road flooding.