Google Earth Engine for Remote Sensing: From Zero to Expert

Discover why Google Earth Engine enables cloud-based, global-scale geospatial analysis with 40+ years of historical imagery, JavaScript and Python APIs, and a code editor.

Explore the Google Earth Engine code editor, a web-based JavaScript API IDE, and learn to manage scripts, assets, and documentation for geospatial analyses and map visualization.

Geographic information systems are defined and two main data types—vector and raster—are explained, covering points, lines, polygons, rasters, and latitude and longitude coordinates, including continuous and discrete rasters.

Explore satellite images and their characteristics, including digital images, pixels, spectral bands (multispectral and hyperspectral), true and false color composites, radiometric resolution, and spatial and temporal resolutions.

Compare active and passive remote sensing sensors and the platforms that carry them, and explain multispectral and hyperspectral imagery across ground, airborne, and satellite platforms for land use classification.

Explore NASA's Landsat program and Landsat 8, with a multispectral and thermal infrared sensor across 11 bands at 15–30 meter resolution, learning atmospheric correction and level 1 and 2 products.

Explore the European Space Agency's Sentinel program, focusing on Sentinel-2 multispectral sensors, data characteristics, processing levels, and their free, climate and land-surface applications.

Explore European Space Agency cloud tools to access Sentinel-2 imagery, compute spectral indices like NDVI in real time, visualize true and false color composites, and download data after login.

Explore Google Earth Engine's vast raster data catalog, including Landsat and Sentinel imagery, elevation and nightlights datasets, and learn to access, preprocess, and classify land use.

Master the basics of JavaScript syntax for Google Earth Engine. Explore variables, strings, lists, dictionaries, comments, and functions to write reusable code.

Discover the basics of JavaScript in Google Earth Engine, declaring numbers, strings, lists, and dictionaries, using var, semicolons, and consistent quotes, with print and console for debugging.

Import satellite data into Google Earth Engine, work with Landsat 8 image collection, filter by region and date, and create a median visualization for mapping study areas.

Explore visualizing Landsat 8 imagery by creating true color (bands 4,3,2) and false color composites, adjusting min/max values, applying color palettes, and centering the map on your area.

Practice JavaScript basics in Google Earth Engine by declaring variables, mapping and reducing collections, and visualizing Landsat images; apply changes to area and time, and create your own image visualizations.

Explore Landsat sensors and data products, including 11 spectral bands from the OLI and TIRS at 15–30 meter resolution, with level one and level two reflectance products for time-series analysis.

Compute single-image calculations in Google Earth Engine, deriving the ndvi from red and near infrared bands using manual and image expression methods, plus visualization and region filtering.

Learn to build a raster composite and compute NDVI in Google Earth Engine using three methods: band-based calculation, built-in NDVI function, and expression, with sensor-specific band selection.

Learn to compute zonal statistics in Google Earth Engine for remote sensing with Landsat data, using a region buffer, masking water, and reducing by zone to obtain mean elevation groups.

Learn to write and map custom JavaScript functions in Google Earth Engine to compute maximum NDVI from Landsat imagery and add it as a new band.

Map a JavaScript function over a Landsat image collection to compute an index such as ndvi, filter by study area and date, and output an image collection for further analysis.

Rename the output image collection by substituting the default day name with a chosen player name, and rename the first image band to in Dubai, then verify in the inspector.

Export images from Google Earth Engine to Google Drive or cloud storage, selecting a region of interest, clipping with a buffer, and configuring scale and projection before running the export.

Import raster and vector assets into Google Earth Engine by using the assets panel, uploading supported formats, and adding them to your script for analysis.

Mosaic Sentinel-2 images and export them as tiffs to Google Drive. Filter by date and geometry, preselect bands, clip to the Paris study area, and reproject for export.

Complete the section six practical task by locating an image or vector data set and importing it as an asset into Google Earth Engine to practice data import and export.

Learn how to work with satellite images for land use and land cover mapping, covering pre-processing steps like cosmetic operations, radiometric calibration, atmospheric and geometric corrections, and subsequent classification.

Iterate over a precipitation dataset in Google Earth Engine to create a binary dry-day mask and track consecutive no-rain days for drought monitoring.

Mask clouds and cloud shadows in sentinel-2 optical imagery using qa60 bands, apply the mask to the image collection, and visualize flood-mapping results over the study area.

Explore flood monitoring with Google Earth Engine by computing the normalized difference water index from Sentinel-2 imagery. Discover two ndwi variants—green–nir and nir–swir—and apply thresholds for water body mapping.

Map floods with Sentinel-2 data and NDWI in Google Earth Engine, mask clouds, compute NDWI, take the maximum, and apply a threshold to produce a water mask.

Apply flood mapping tasks in Google Earth Engine by changing sentinel periods, testing NDWI thresholds, and comparing band configurations to identify the best threshold and bands for your study area.

Learn what machine learning is and how data driven algorithms learn from labeled data to power classification and regression, covering supervised, unsupervised, semi supervised, and reinforcement learning, with deep learning.

Explore the supervised land use and land cover mapping workflow, from defining mutually exclusive and exhaustive classes to collecting representative training data and assessing accuracy for map outputs.

Apply image classification in Google Earth Engine by adjusting study region to recent dates, and experiment with a cluster algorithm, number of classes, and training pixels to find best option.

Map linear regression on a two-band nighttime light image collection in Google Earth Engine, with time as the independent variable, to reveal per-pixel trends.

Explore js, remote sensing, data science, and machine learning through the instructor’s Udemy page, YouTube Geo World channel, and social media updates that guide course selection and promotions.