Core GIS : Land Use Land Cover & Change Detection in QGIS

Explore how digital images are built from pixels with single or multiple spectral bands and how spatial, spectral, radiometric, and temporal resolutions shape land use and change detection in QGIS.

Explore remote sensing sensors and platforms, distinguishing passive optical, active radar and lidar systems, with multispectral and hyperspectral imagery from Landsat and Sentinel on ground, airborne, and satellite platforms.

Explore the basics of remote sensing for land use land cover mapping and change detection using Landsat imagery and spectral signatures.

Learn to download satellite images for land use mapping with the semi-automatic classification plugin in QGIS, sourcing Landsat and Sentinel data from USGS, NASA, and Copernicus.

Apply remote sensing image preprocessing to prepare satellite data for land use and land cover mapping. Learn radiometric, atmospheric, and geometric correction, plus cloud masking and image quality considerations.

Learn to create a multilayer layer stack from Landsat 8 bands in QGIS. Then visualize true color and false color composites to highlight vegetation and deforestation.

Perform radiometric and atmospheric correction on Landsat 8 imagery to convert digital numbers to reflectance, enabling classification with the semi-automatic classification plugin that uses dark object subtraction.

Discover remote sensing image sources for LULC mapping, including color imagery (red, green, blue) and data beyond the visible range, via Landsat, Sentinel, Copernicus, Google Earth tools, and commercial providers.

Introduce digital image classification as the main step to convert Landsat satellite imagery into a land use land cover map, using spectral patterns, training data, and supervised or unsupervised approaches.

Explore supervised and unsupervised image classification for land use and land cover, including training samples, interpretation keys, clustering, and using reference data to assign identities.

Load a Sentinel-2 multi-band image and run unsupervised classification with the semi automatic classification plugin, using ISODATA or K-means clustering and a chosen band set.

Explore the stages of supervised land use and land cover classification: define mutually exclusive and exhaustive classes, collect training data, compute spectral statistics, classify pixels, and assess accuracy for maps.

Create a QGIS project, load Landsat 5 imagery from 1985, apply dark object subtraction atmospheric correction, and compile data for vegetation, soil, and water using polygons with semi-automatic classification plugin.

Explore common image classification algorithms for land use and land cover in GIS, including supervised and unsupervised methods like minimum distance to means, maximum likelihood, and random forest.

Learn land use and land cover classification with the semi-automatic classification plugin using spectral angle mapping, build training data for forest, soil, and water, and preview and refine results.

Create land use land cover maps from Landsat 5 using the maximum likelihood algorithm with provided training data, then review vegetation, soil, and water classifications.

Assess accuracy of land use land cover maps using reference data and confusion matrices, and compute overall, user, and producer accuracies from the validation results.

Create validation data for vegetation, soil, and water to perform numeric accuracy assessment of land use land cover maps in QGIS using a semi-automatic classification plugin and validation samples.

Learn to perform accuracy assessment on land use land cover maps using a validation dataset and the semi-automatic classification plugin in QGIS, interpreting producer, user, and overall accuracy.

Map land use and land cover with Landsat 8 imagery from 2013 in Brazil using dark object subtraction atmospheric correction, training data, and three semi automatic classifiers, then compare accuracy.

Import a raster from random forest or SVM classification into QGIS, then visualize a five-class land cover map and customize colors and labels for clear reporting.

Create a land use land cover map from your final image classification in QGIS by adding a map to a print layout and exporting as tiff or pdf.

Learn change detection in QGIS by comparing temporal multispectral imagery to create land use land cover maps, analyze urban sprawl and vegetation change with spectral indices.

Demonstrate change detection in QGIS by comparing 1985 Landsat 5 forest, soil, and water with 2013 Landsat 8 classifications; interpret change codes, compute pixel sums and areas, and relabel outputs.

Create a map from change detection results in QGIS by configuring a print layout, adding a map, legend, scale bar, title, and north arrow; export as image and save project.

Explore remote sensing, data science, and machine learning through curated Udemy courses and free Jio World videos, with regular updates and guided learning paths.