Remote Sensing Fundamentals Training Course

Participants can match sensor platforms and imagery specifications to defined information requirements.

• Satellite, airborne, UAV and ground platforms matched to collection constraints
• Multispectral, hyperspectral, thermal and radar sensor characteristics compared
• Orbital revisit cycles, viewing geometry and sun angle effects
• Sentinel, Landsat and commercial imagery selection criteria for project needs
• UAV acquisition limitations: weather, regulations, battery capacity and permissions
• Sensor specifications translated into minimum data requirements documents

Participants can manage coordinate reference systems and raster alignment for spatially reliable analysis.

• Coordinate reference systems, datums and map projections for imagery alignment
• Geographic and projected coordinates distinguished using EPSG identifiers
• UTM zone selection and hemisphere conventions for regional mapping
• Reprojection risks assessed through extent, units and distortion patterns
• Ground control point distribution and residual error principles
• Raster cell size, extent and alignment controlled in QGIS

Participants can locate, assess and document imagery suitable for an analysis request.

• STAC catalogues, Copernicus Data Space and USGS EarthExplorer search methods
• Acquisition dates filtered against phenology, cloud cover and event timing
• Scene metadata inspected for processing level, bands and quality flags
• ISO 19115 metadata elements recorded for imagery provenance
• Licensing terms checked for redistribution, derivatives and commercial use
• Data inventory structured with source, date, resolution and intended analysis

Participants can organise, inspect and display raster imagery within a structured QGIS project.

• QGIS interface, Browser panel and layer properties used efficiently
• Raster and vector layers organised using consistent project conventions
• Band combinations rendered with contrast stretches and colour ramps
• Identify tools and pixel queries used to inspect image values
• Processing Toolbox operations documented through reproducible model history
• GeoPackage project structure applied to inputs, outputs and reference layers

Participants can prepare imagery for analysis by applying appropriate radiometric, atmospheric and masking procedures.

• Level-1 and Level-2 products distinguished by correction status
• Radiometric calibration converts digital numbers into physical measurement values
• Atmospheric correction purpose, assumptions and limitations evaluated
• Cloud, shadow and cirrus masks applied using quality assessment bands
• Image mosaics built with overlap checks and radiometric consistency
• Clipping, resampling and band stacking prepared for analysis

Participants can assess positional quality and register imagery for multi-date comparison.

• Georeferencing workflow uses control points, transformation models and resampling
• Affine, polynomial and projective transformations matched to source distortion
• Nearest neighbour, bilinear and cubic resampling effects compared
• Root mean square error interpreted alongside independent positional checks
• Orthorectification purpose explained using terrain displacement and sensor geometry
• Co-registration accuracy verified before multi-date change analysis

Participants can identify and delineate land-surface features using systematic visual interpretation methods.

• Image interpretation elements: tone, texture, pattern, shape, shadow and association
• False-colour composites selected for vegetation, water and urban feature discrimination
• Interpretation keys designed with reference examples and decision rules
• Ancillary maps, field observations and high-resolution basemaps corroborate labels
• Digitising protocols define scale, topology, attributes and minimum mapping units
• Feature confidence recorded using evidence source and interpretation certainty

Participants can calculate and interpret spectral indices for vegetation, water and built-up land analysis.

• Band arithmetic calculates normalized difference indices from calibrated reflectance
• NDVI interpretation considers canopy density, soil background and saturation
• NDWI and MNDWI selected for open-water delineation conditions
• NDBI and built-up indices assessed against bare-soil confusion
• Raster calculator expressions documented with band names and scaling factors
• Index thresholds derived from reference samples and visual inspection

Participants can create supervised land-cover classifications using labelled samples and documented class definitions.

• Classification problem framed through classes, mapping scale and decision criteria
• Training samples designed for representativeness, purity and class separability
• Spectral feature selection combines bands, indices, texture and elevation
• Random Forest classification parameters set using labelled reference data
• Unsupervised clustering contrasted with supervised land-cover classification
• Class schema aligned with project questions and stakeholder definitions

Participants can compare pixel, object and rule-based methods for spatial feature mapping.

• Pixel-based and object-based classification approaches compared for spatial context
• Image segmentation parameters control object scale, shape and compactness
• Texture measures distinguish crops, roofs and dense urban surfaces
• Rule-based classification combines spectral thresholds, objects and contextual criteria
• Post-classification filtering removes isolated pixels without concealing meaningful features
• Class recoding and aggregation support reporting at required thematic levels

Participants can evaluate classification quality and report the uncertainty affecting map-based decisions.

• Reference sample design uses stratification, randomisation and adequate sample sizes
• Confusion matrices calculate producer's, user's and overall accuracy
• Precision, recall and F1 score interpreted for thematic classes
• Area-adjusted accuracy estimation accounts for unequal sampling probabilities
• Positional accuracy checks compare mapped boundaries against independent reference data
• Uncertainty statements connect error sources, confidence and decision consequences

Participants can identify, measure and report land-surface change using multi-date imagery.

• Post-classification comparison identifies transitions between independently classified dates
• Image differencing detects continuous change using matched sensor measurements
• Change thresholds calibrated against stable areas and reference observations
• Multi-temporal compositing reduces cloud effects and seasonal noise
• Time-series charts reveal trends, anomalies and abrupt land-cover shifts
• Change matrices report gains, losses, persistence and transition pathways

Participants can derive terrain variables and use them with imagery for spatial screening.

• Digital elevation models represent terrain as regularly spaced raster surfaces
• Slope, aspect, hillshade and contours derived using terrain analysis tools
• Hydrological conditioning fills sinks before flow-direction and watershed modelling
• Viewshed analysis evaluates line-of-sight constraints from specified observer locations
• Elevation datum and vertical accuracy assessed before engineering interpretation
• Terrain derivatives combined with imagery for landslide and flood screening

Participants can interpret the uses and limitations of thermal and radar imagery.

• Thermal infrared measurements relate emitted radiance to land surface temperature
• Emissivity, atmospheric effects and acquisition timing constrain thermal interpretation
• SAR backscatter responds to roughness, moisture, geometry and wavelength
• Speckle reduction and terrain correction prepare radar images for analysis
• Sentinel-1 applications support flood mapping beneath persistent cloud cover
• Optical and radar evidence combined for complementary land-surface observations

Participants can integrate remote sensing outputs with GIS data and produce decision-ready maps.

• Zonal statistics summarise raster values within administrative or management polygons
• Raster-vector overlays link classification outputs with assets, populations and boundaries
• Area calculations use equal-area projections and documented class definitions
• Map layouts communicate purpose, scale, uncertainty, sources and production date
• Symbology choices distinguish measured values, classified categories and reference features
• OGC WMS and WCS concepts support imagery service publication

Participants can plan a remote sensing project with defined requirements, controls and reporting criteria.

• Problem statements define decisions, users, spatial extent and required evidence
• Data specifications set sensors, dates, resolutions, accuracy and licensing constraints
• Processing plans sequence acquisition, preprocessing, analysis, validation and reporting
• Quality assurance checkpoints document parameters, versions and reviewer decisions
• Ethical considerations address surveillance, sensitive locations, consent and data minimisation
• Project risks cover cloud cover, access, bias, cost and schedule

Participants can build, review and present a complete documented remote sensing analysis for a defined decision question.

• Project brief translated into a defensible remote sensing analysis design
• Imagery, reference data and processing choices justified against stated requirements
• Reproducible QGIS workflow produces mapped outputs and supporting evidence
• Accuracy results and limitations documented for intended decision-makers
• Final map layout applies cartographic hierarchy, metadata and uncertainty notes
• Peer review findings incorporated before presentation of methods and conclusions