Session Descriptions
Analysis to Impact
Presenter: Chris Herwig
Description: This session will focus on tools and technologies to amplify the impact of your scientific analyses and research. Thinking about publishing the results of your analysis? This session is right for you. We'll highlight some exciting technologies you can take advantage of today to help get your analysis out of the Code Editor and into the world. We'll touch on a variety of tools, including Google Earth and Maps in the context of communicating your message on the rich backdrop of imagery and other geo data. Beyond mapping, we're a Search company after all, so we'll also go over content markup you can use to make your content more discoverable and semantics more explicit -- so it's easier to find and use by other researchers.
Prerequisites: None
Content: Slides
Basics of Google Maps Platform (Maps API)
Presenter: Christiaan Adams
Description: Take "baby steps" into the world of Google Maps Platform by trying out the Google Maps Javascript API. In this session, you'll create a basic "Hello World" Maps API webpage, and will learn how to do things like change the options (background map type, initial location, etc.), overlay data and KML layers, and start to showcase your first Earth Engine maps. While this is taught for non-programmers by a non-programmer, we will work with snippets of code copied-and-pasted from the developers.google.com website.
Prerequisites: Some comfort opening the code of a basic HTML page, but no prior JavaScript coding necessary.
Content: Slides
Building Scalable (static) Websites from Earth Engine data
Presenter: Ollie Guinan
Description: Learn to use the latest cloud technologies to publish a website that lets users explore complex geospatial data from Earth Engine, but still performs fast and scales well for high traffic days. You will learn how to publish analysis as layers to the Google Maps Platform (Maps API), measure the performance of your website, add scalable caching and traffic management features so that all of your users enjoy a smoothly running site. We will provide an example that you can extend and use in your own work.
Prerequisites: Basic understanding of web servers and APIs. Some Earth Engine coding experience. No deep coding or cloud knowledge necessary.
Content: Slides
Building User Interfaces in Earth Engine (EE103)
Presenter: Lauren Scott
Description: This class will introduce new tools that make it easy to design interactive Earth Engine visualizations. Learn to make interactive maps with Earth Engine, including widgets for labels, buttons, sliders, panels, charts and maps.
Prerequisites: None
Conservation workflows in Earth Engine
Presenter: Dave Thau
Description: Interested in using Earth Engine for conservation work? This is the session for you. Thau will give a quick overview of conservation-related efforts and datasets in Earth Engine. We'll have summit participants from Conservation International, Defenders of Wildlife, UNEP-WCMC, and World Wildlife Fund present the work they've been doing with Earth Engine. And we'll have plenty of time to discuss other Earth Engine applications in conservation, and to brainstorm about useful datasets and algorithms.
Prerequisites: None.
Content:
- EEUS 2018 Conservation Session Intro Slides
- Conservation International - Using the Cloud to Support Conservation on the Ground
- Defenders of Wildlife - Monitoring the U.S. Endangered Species Act
- World Conservation Monitoring Centre (WCMC) - EE for Conservation: Highs, lows and ways to go
- World Wide Fund for Nature (WWF) - Satellite Remote Sensing to Save Species
Earth Engine in India!
Presenter: Allie Lieber
Description: We officially launched Geo for Good in India this year! As part of the launch we went on a 15-stop roadshow around the country, teaching Earth Engine to graduate students, professors, and scientists. We will share our takeaways and hope to hear your thoughts about how to use Earth Engine to make the most positive impact in India.
Prerequisites: None
Content: Slides
EE101A: Introduction to Earth Engine for beginners, part A
Presenter: Dave Thau
Description: The Earth Engine API (application programming interface) provides the ability to create your own algorithms to process raster and vector imagery. This session is geared toward people who would like to analyze satellite and vector data. The session will be hands-on, using the Earth Engine Javascript code editor. This part of the class will focus on accessing imagery, creating composites, and running analyses over stacks of images.
Prerequisites: Familiarity with at least one software language, or at least not be afraid of learning as we go.
Content: Slides
EE101B: Introduction to Earth Engine for beginners, part B
Presenter: Dave Thau
Description: This is part two of the Getting Started class. This part of the class will continue where part 1 left off, focusing on computing statistics on imagery, creating charts, exporting the results of your analyses.
Prerequisites: Earth Engine 101A.
Content: Slides
EE102: Supervised Classification on Time-series
Presenter: Noel Gorelick
Description: Understand the steps for creating, using and validating a classification in Earth Engine and the basic limitations of the classifier system (table size, tileScale, etc). Learn how to do parameter optimization. Learn how to do inter-annual compositing and investigate the effects of multi-temporal feature vectors and multi-sensor fusion on classifier accuracy.
Prerequisites: EE101 or equivalent
Content: Code Repository
Hands On: Change Detection from Landsat composites
Presenter: Nick Clinton
Description: A hands-on course in change detection from Landsat annual composites. Methods include post classification comparison, direct classification of change, spectral index differencing, spectral angle differencing, spectral distance thresholds.
Prerequisites: Earth Engine 101 or equivalent.
Content: Slides, Code Repository
How to be an Earth Engine trainer
Presenter: Nick Clinton
Description: Interested in leading your own Earth Engine trainings? Get an overview of the materials and methods the Earth Engine team uses to conduct trainings all over the world.
Prerequisites: Earth Engine 101 and 201 or equivalent.
Content: Slides
Image segmentation and object based methods for image analysis
Presenter: Noel Gorelick
Prerequisites: None
Content: Video, Slides, Code Repository
Importing and Exporting data to/from Earth Engine
Presenter: Chris Herwig
Description: This session will cover how users can bring their own vector/raster data into Earth Engine and the different ways users can take the results of their analyses out of Earth Engine.
Prerequisites: None
Content: Slides
Ingestion and the public data catalog
Presenter: Simon Ilyushchenko
Description: Are you struggling to ingest data into Earth Engine? Have you been waiting forever for the EE team to ingest new datasets? This session will go over common problems that can happen when uploading rasters and vectors. We will also discuss possible new ways of adding more datasets into the EE catalog without sacrificing data quality. For example, can external groups curate parts of the catalog?
Prerequisites: None
Content: None (discussion only)
Introduction to TensorFlow models and Earth Engine
Presenter: Nick Clinton and/or Chris Brown
Description: Introduction to coding with the TensorFlow Estimator API. This course is an end-to-end walkthrough of generating training and validation data in Earth Engine, exporting to the TFRecord format, building a simple DNN model in TensorFlow, training and evaluation with the EE data, classification of an image as a TFRecord file, and import back to Earth Engine.
Prerequisites: Python programming. Google Colab is installed as a Drive service.
Landcover Estimation and Validation*
Presenter: Pontus Oloffson
Description: Classification errors in remote sensing-based maps may introduce severe bias in mapped areas of land cover and land cover change. To accommodate such errors, sample-based unbiased estimation of area and map accuracy is required but collecting sample data is often costly and time consuming. A solution that provides access to reference data for any location in the world without having to download and process the data would be game-changer – Google Earth Engine has the potential to provide such a solution, which we will discuss in this session.
Prerequisites: EE101 or equivalent.
Content: Slides
Monitoring with Breakpoint Algorithms*
Presenter: Robert Kennedy
Description: Thanks to the hard work of the GEE team and collaborators, you now have the power to tap image time series archives with change detection algorithms based on temporal segmentation. Here, we will give a primer in LandTrendr, VeRDET, and EWMACD. We’ll cover strategies to build appropriate image collections to feed the algorithms, how to determine best approaches to examine parameter impacts, and the basics of making sense of the outputs.
Prerequisites: EE101 or equivalent.
Content: Slides
NexGenMap*
Presenter: Tasso Azevedo and Leandro Parente
Description: The NextGenMap is a project that aims establish the next generation of land use and land cover mapping and monitoring tools, using high-res images (i.e. Planet data), cloud-computing platforms (i.e. Google Earth Engine and Google Cloud Computing) and machine learning and deep-learning techniques (e.g. Random Forest, U-Net and LSTM). This session will present the context and contributions of the project (e.g. Planet weekly mosaic and end-to-end classification approaches) and will have a hands-on section, applied to the detection of forest, pasture and other land uses and objects.
Prerequisites: Basic Javascript and Python programming skills
Overview of Earth Engine in the SERVIR program*
Presenter: Dave Saah
Description: SERVIR is a unique partnership between the U.S. Agency for International Development (USAID) and the U.S. National Aeronautics and Space Agency (NASA) focused on bringing space-based technologies to environmental decision makers in developing regions. As a geospatial data-for-development program, SERVIR works in partnership with leading regional organizations to help countries use information provided by Earth observing satellites and geospatial technologies to manage climate risks and improve disaster preparedness. During this session, we will present some of our SERVIR Earth Engine based decision support tools and outreach.
Prerequisites: None
Overview of the accessibility dataset*
Presenters: Dan Weiss and Harry Gibson (Oxford University)
Description: The creators of the Accessibility Map (nature.com/articles/nature25181) will walk you through developing your own map using their underlying friction surface and least cost path functionality. This will be done in both Earth Engine and R.
Prerequisites: None
Content: Slides
Overview of the Google Cloud Platform
Presenter: Matt Hancher
Description: Learn how to use Earth Engine together with other products in the Google Cloud Platform. This talk will introduce you to a variety of Cloud Platform services, including Cloud Storage for managing and serving data, Compute Engine and Dataflow for running computations, and Pub/Sub for building real-time analysis applications. Then we'll show you how you can seamlessly use these systems together with Earth Engine to build powerful geo data processing applications in the cloud.
Prerequisites: This class is aimed at experienced Earth Engine users who have a general familiarity with cloud computing concepts, but no prior experience with Google Cloud Platform is assumed.
Content: Slides
Running Earth Engine with Node.js
Presenter: Don McCurdy
Description: Learn how to use the Earth Engine JavaScript API outside of the Code Editor, with Node.js. This session will discuss moving your Earth Engine scripts into command line tools, App Engine applications, Cloud Functions, and more.
Prerequisites: Earth Engine 101 or prior experience in using Earth Engine.
Content: Slides
The Earth Engine Python Environment (EE104)
Presenter: Tyler Erickson
Description: This session will introduce using tools from the open source Jupyter project (http://jupyter.org) and Python visualization libraries to develop algorithms with the Earth Engine Python API.
Prerequisites: 1) Earth Engine 101 or prior experience in using Earth Engine. 2) A GitHub account (https://github.com/join).
Content: Slides