Geospatial Data Analysis and Remote Sensing: 5 Courses in 1

Choose stable long-term releases for qgis to avoid experimental builds. Compare current versions and interfaces across windows, mac, linux, and android to ensure compatibility in your geospatial workflow.

Learn how to select and install stable QGis versions and older releases. Manage plugins like semi automatic classification plugin and KSP plugin, and ensure compatibility for remote sensing analyses.

install qgis on windows by downloading from qgis.org, choose latest or long-term release, select 64-bit, complete installation, and launch desktop version (with grass integration) for geospatial analysis.

Explore the semi-automatic classification plugin for QGIS to perform supervised classification, reflectance conversion and preprocessing, training data creation (points, polygons, or ROIs), change detection, and accuracy assessment.

Learn to customize the QGIS interface by installing and managing plugins, exploring community plugins like sentinel and semiautomatic classification, and configuring tools for remote sensing workflows.

Explore the QGIS interface by opening existing projects, managing layers in the layer panel, and customizing panels and toolbars; save projects and use identify tools with coordinates.

Install or fire tool box on your computer and activate it to apply machine learning algorithms for image classification, following a clear step-by-step guide and panel access to toolbox tools.

Sign up for Google Earth Engine to access cloud-based earth science data for classification, using the code editor and Landsat time-lapse animations; a Gmail account is required.

Define the main principles of GIS and its four core ideas—create, manage, analyze, display—using maps, geo data, spatial analysis, and apps to work with location.

Explore geodata foundations for geospatial analysis, covering vector data with points, lines, and polygons anchored to latitude and longitude, and raster data defined by pixels with continuous and discrete values.

Set up your PC for geospatial analysis by selecting hardware requirements and exploring top open source GIS tools like QGIS, GRASS GIS, SAGA GIS, and Whitebox GAT.

Explore sensors and platforms, contrasting active radar and LiDAR with passive sensors like Sentinel and Landsat, and using multispectral and hyperspectral data for land use and land cover classification.

Examine digital images, pixels, and pixel values, then differentiate spatial, spectral, radiometric, and temporal resolutions in multispectral and hyperspectral data.

Learn to download satellite imagery for study areas using the semiautomatic classification plugin in QGIS, including setting up accounts and selecting coordinates, dates, and cloud-free Landsat products.

Learn to build a multilayer Landsat 8 stack in Kuji and create true color and false color composites to visualize vegetation in the Amazon rainforest.

Perform radiometric and atmospheric correction on Landsat 8 imagery, converting digital numbers to reflectance with a semi-automatic classification plugin and applying dark object subtraction; create a multi-band layer stack.

Access a range of remote sensing data sources for imagery download and analysis, including Google Earth Engine, Landsat Explorer, Sentinel Hub, Earth Explorer, and Copernicus Open Access Hub.

Explore how machine learning uses data-driven algorithms to learn from data. Identify the main types: supervised, semi supervised, unsupervised, and reinforcement learning, and how labeling shapes predictions.

Learn land use and land cover mapping and change detection with open source tools like QGIS, Google Earth Engine, and SNAP, using supervised, unsupervised, and object-based classifications.

Examine supervised and unsupervised image classification for land use, using training data to label spectral attributes or uncover natural spectral clusters with clustering algorithms such as k-means.

Perform unsupervised k-means image analysis in QGIS to classify a city image into three classes: water, vegetation, and built-up, with color-coded results saved to an output file.

define mutually exclusive and exhaustive land use and land cover classes, collect representative training data, and classify pixels to produce a land use map, followed by accuracy assessment.

Explore a range of supervised and unsupervised image classification algorithms, including minimum distance to mean, maximum likelihood, decision trees, random forest, and support vector machines, with examples and accuracy considerations.

Use the minimum distance classification on a three-class land use and land cover dataset (forests, soil, water) and compare with maximum likelihood, noting visual plausibility and upcoming accuracy assessment.

Map land use land cover from a 2013 Landsat 8 image of Brazil using dark-object subtraction atmospheric correction, prepare training and validation data, classify with three algorithms, and assess accuracy.

Create a change-detection map in QGIS by building a print layout with a map, legend, scale bar, north arrow, and title, then save the map and project.

Learn to import Sentinel-2 data with Google Earth Engine, define a Dubai geometry, filter by date, and visualize images in true color and false color composites.

Learn unsupervised image classification with k-means in Google Earth Engine for Dubai, creating training data, running clustering into five to ten classes, and exporting results for visualization and further analysis.

Learn to perform supervised classification in Google Earth Engine with sentinel imagery and a random forest classifier, creating training data for water, vegetation, burn, sand, and rocks.

Apply supervised classification on the cloud using Google Earth Engine Explorer to a chosen geographic area. Create training data, assess map accuracy, and export a TIFF for use in QGIS.