About This Course
Remote sensing technology has transformed the way organisations monitor land cover, manage natural resources, assess crop health, and respond to environmental change. As satellite constellations multiply and sensor capabilities advance, the volume and complexity of geospatial data available to analysts has grown exponentially — demanding a higher level of technical proficiency than ever before.
This Advanced Remote Sensing Data Analysis Training programme is designed for professionals who already possess a foundational understanding of GIS and remote sensing concepts and are ready to move into sophisticated analytical workflows. Over ten intensive days, participants will work directly with multispectral, hyperspectral, SAR, and LiDAR datasets, applying classification algorithms, change detection techniques, time-series analysis, and machine learning models to derive meaningful insights from raw imagery.
Delivered through a blend of structured instruction, software-based practical sessions, and real-world case studies drawn from agriculture, forestry, hydrology, and urban environments, the programme is designed to produce analysts who can confidently design and execute end-to-end remote sensing workflows. Participants will work in leading platforms including QGIS, ENVI, Google Earth Engine, and Python-based geospatial libraries.
By the end of this programme, participants will leave with a portfolio of completed analytical projects, a deep command of advanced processing techniques, and the professional confidence to lead remote sensing initiatives within their organisations.
Who Should Attend
- GIS Analysts and Geospatial Scientists seeking to upgrade their technical remote sensing skills
- Environmental Scientists and Ecologists working with satellite or aerial imagery
- Agriculture and Food Security Analysts using earth observation data for crop monitoring
- Forestry and Land Use Planners requiring advanced change detection capabilities
- Water Resource and Hydrology Specialists integrating remote sensing into catchment analysis
- Natural Resource Management Officers in government agencies and conservation bodies
- Research Scientists and Postgraduate Researchers in earth observation disciplines
- Urban and Regional Planners using satellite data for infrastructure and growth monitoring
- Climate Change and Disaster Risk Reduction Professionals needing satellite-based situational awareness
Learning Objectives
- Understand the physical principles of electromagnetic radiation and sensor design underpinning advanced remote sensing systems
- Acquire, pre-process, and radiometrically correct multispectral, hyperspectral, LiDAR, and SAR datasets
- Apply supervised and unsupervised classification algorithms including Support Vector Machines, Random Forest, and neural network classifiers
- Perform multi-temporal and time-series analysis to detect land cover change and seasonal vegetation dynamics
- Derive and interpret advanced spectral indices for vegetation health, soil moisture, water quality, and urban heat
- Process and analyse Synthetic Aperture Radar (SAR) data for surface deformation, flood mapping, and biomass estimation
- Extract and model 3D surface information from LiDAR point clouds and Digital Elevation Models
- Automate remote sensing workflows using Python scripting with libraries including GDAL, Rasterio, and scikit-learn
- Conduct accuracy assessments and apply validation frameworks to remote sensing classification outputs
- Integrate Google Earth Engine for large-scale, cloud-based geospatial analysis and visualisation
- Develop field data collection strategies to support and validate satellite-based analysis
- Produce professional-quality cartographic outputs, reports, and data products for stakeholder communication
- Design and implement an end-to-end remote sensing project addressing a real-world environmental or agricultural challenge
Course Modules
Module 1: Foundations of Advanced Remote Sensing Systems
- Electromagnetic spectrum interactions with vegetation, soil, water, and built surfaces
- Sensor types and design trade-offs: spatial, spectral, temporal, and radiometric resolution
- Overview of key satellite missions: Sentinel series, Landsat-8/9, MODIS, ALOS-2, and commercial constellations
- Data access portals and download strategies: Copernicus, USGS EarthExplorer, and Planet Labs
- Orbital mechanics and acquisition geometry: sun-synchronous orbits, swath width, and revisit frequency implications for analysis
- Metadata interpretation: understanding scene identifiers, acquisition parameters, and quality flags across major data providers
Module 2: Image Pre-Processing and Radiometric Correction
- Geometric correction, orthorectification, and image co-registration techniques
- Atmospheric correction methods: DOS, FLAASH, Sen2Cor, and 6S model application
- Radiometric calibration: converting digital numbers to top-of-atmosphere and surface reflectance
- Cloud and shadow masking strategies for optical imagery pre-processing pipelines
- Bidirectional Reflectance Distribution Function (BRDF) correction and its importance for multi-date and multi-sensor compositing
- Pan-sharpening methods: Gram-Schmidt, IHS, and Brovey transforms for resolution enhancement while preserving spectral fidelity
Module 3: Multispectral Image Analysis and Spectral Indices
- Band combination selection for vegetation, geology, urban, and water feature enhancement
- Deriving and interpreting NDVI, EVI, NDWI, MNDWI, NDBI, and SAVI indices
- Principal Component Analysis (PCA) and Tasselled Cap transformation for data dimensionality reduction
- Spectral profiling and endmember identification for land cover characterisation
- Linear spectral unmixing: estimating sub-pixel fractional cover of vegetation, bare soil, and impervious surfaces
- Index saturation effects and sensor-specific calibration adjustments for high-biomass and dense urban environments
Module 4: Hyperspectral Remote Sensing and Analysis
- Principles of hyperspectral imaging and differences from multispectral data
- Hyperspectral data formats and pre-processing: noise reduction, smile correction, and spectral smoothing
- Spectral libraries and spectral angle mapper (SAM) classification for material identification
- Applications in precision agriculture, mineral mapping, and vegetation stress detection
- Minimum Noise Fraction (MNF) transformation for dimensionality reduction and signal-to-noise enhancement in hyperspectral cubes
- Continuum removal and derivative spectroscopy techniques for isolating diagnostic absorption features in target spectra
Module 5: Image Classification: Supervised and Unsupervised Methods
- Unsupervised classification: K-Means and ISODATA clustering and cluster labelling strategies
- Supervised classification: Maximum Likelihood, Support Vector Machines, and Random Forest algorithms
- Training sample design: sample size, spatial representativeness, and class separability assessment
- Post-classification refinement: majority filtering, sieve, and contextual editing techniques
- Object-based image analysis (OBIA): segmentation parameters, shape and spectral attributes, and rule-set classification workflows
- Handling class imbalance in training datasets: oversampling, undersampling, and weighted loss strategies for rare land cover classes
Module 6: Machine Learning and Deep Learning for Remote Sensing
- Feature engineering from spectral bands, indices, and texture metrics for classifier input
- Random Forest and Gradient Boosting classifiers implemented in scikit-learn on raster datasets
- Convolutional Neural Networks (CNNs) for image segmentation and object detection in satellite imagery
- Model training, hyperparameter tuning, and overfitting management in geospatial ML workflows
- Transfer learning with pre-trained CNN architectures: adapting ImageNet-trained models to multispectral remote sensing classification tasks
- Spatial cross-validation strategies for geospatial ML: blocking by geography to prevent data leakage and produce honest accuracy estimates
Module 7: SAR Data Processing and Applications
- SAR imaging geometry: backscatter, polarisation, and coherence fundamentals
- SAR pre-processing in SNAP: orbit correction, calibration, speckle filtering, and terrain correction
- SAR applications: flood extent mapping, crop type discrimination, and aboveground biomass estimation
- Differential InSAR (DInSAR) for ground subsidence and surface deformation monitoring
- Polarimetric SAR decomposition: Pauli and Freeman-Durden decompositions for separating surface, double-bounce, and volume scattering mechanisms
- SAR time-series coherence analysis for infrastructure mapping and monitoring urban construction activity over multi-temporal stacks
Module 8: LiDAR Data Processing and 3D Analysis
- LiDAR acquisition principles: discrete return versus full-waveform sensors and point density considerations
- Point cloud processing in CloudCompare and LAStools: filtering, ground classification, and normalisation
- Deriving Digital Terrain Models, Digital Surface Models, and Canopy Height Models from LiDAR
- Forest structure metrics: basal area, crown volume, and tree height extraction for biomass modelling
- Individual tree segmentation algorithms: watershed and region-growing approaches applied to normalised point clouds and CHM rasters
- Fusing LiDAR-derived structural metrics with multispectral or hyperspectral data to improve species classification and carbon stock estimation
Module 9: Change Detection and Time-Series Analysis
- Image differencing, ratioing, and post-classification comparison methods for binary change maps
- Multi-temporal compositing and harmonic analysis for seasonal vegetation trend detection
- LandTrendr and BFAST algorithms for detecting disturbance and recovery in time-series datasets
- GEE-based time-series analysis: building and exporting pixel-level change trajectories from Landsat archives
- Continuous Change Detection and Classification (CCDC) for near-real-time monitoring of land surface transitions using dense time stacks
- Attributing detected changes to specific drivers: separating deforestation, fire, drought stress, and phenological shifts using ancillary datasets
Module 10: Accuracy Assessment and Validation Frameworks
- Designing statistically valid sampling schemes: stratified random, systematic, and cluster sampling
- Constructing confusion matrices and computing overall accuracy, Kappa coefficient, and F1-score
- Area estimation with uncertainty bounds using good practice guidance from Olofsson et al. 2014
- Field data collection planning: GPS protocols, data entry standards, and reference data management
- Interpreter variability and inter-rater agreement assessment: managing bias and consistency in reference data labelling for validation
- Validation of continuous outputs such as biomass and canopy height maps: applying RMSE, bias, and relative error metrics with appropriate reference datasets
Module 11: Google Earth Engine for Large-Scale Analysis
- GEE platform architecture: image collections, feature collections, and the JavaScript and Python APIs
- Building scalable spectral index time-series and multi-year composites across continental extents
- Applying machine learning classifiers within GEE for national-scale land cover mapping
- Exporting results, sharing assets, and publishing interactive maps through GEE Apps
- Memory and computation limits in GEE: tiling strategies, pyramid policies, and best practices for optimising large-scale processing tasks
- Integrating external datasets with GEE: ingesting custom rasters and vector assets, and joining GEE outputs with tabular data in Earth Engine
Module 12: Python Automation for Remote Sensing Workflows
- Raster data manipulation with GDAL and Rasterio: reading, writing, reprojecting, and masking
- Automated batch processing pipelines for image download, pre-processing, and index calculation
- Integrating geopandas and shapely for vector-raster operations and zonal statistics extraction
- Building reproducible analysis notebooks and packaging scripts for team-level reuse
- Parallelising raster processing tasks using concurrent.futures and Dask for handling large multi-scene datasets efficiently
- Version control and environment management for remote sensing projects: structuring repositories with Git and managing dependencies with conda environments
Module 13: Capstone Project and Professional Communication of Results
- Capstone project design: problem definition, dataset selection, and analytical approach planning
- End-to-end workflow execution: pre-processing through classification, validation, and change analysis
- Cartographic design principles for professional remote sensing map products and technical reports
- Presenting findings to non-specialist stakeholders: translating analytical outputs into policy-relevant narratives
- Documenting analytical workflows for reproducibility: writing methods sections that satisfy peer-review and technical audit standards
- Critical self-evaluation of project outputs: identifying sources of error, articulating limitations, and proposing defensible directions for further analysis
Requirements & Prerequisites
Participants should have a working knowledge of GIS principles and at least one GIS platform such as QGIS or ArcGIS, along with foundational familiarity with remote sensing concepts such as band combinations, image classification, and basic spectral analysis. Prior exposure to Python or any scripting language is beneficial but not mandatory, as the programming sessions are structured to build from fundamentals. Participants are expected to bring a laptop with QGIS and Anaconda (Python) pre-installed prior to the first day of training.
Training Methodology
- Instructor-led conceptual sessions using annotated satellite imagery and real sensor specification sheets to ground theoretical principles in practical context
- Software laboratory sessions in QGIS, ENVI, and Google Earth Engine where participants process authentic datasets from Sentinel-2, Landsat, ALOS PALSAR, and ICESat-2 missions
- Step-by-step Python scripting workshops using Jupyter Notebooks, with progressively complex exercises in GDAL, Rasterio, and scikit-learn applied to geospatial raster data
- Case study analysis drawn from documented projects in East African agriculture, Sub-Saharan forest monitoring, and Arabian Peninsula urban expansion
- Peer-review exercises in which participants critique each other's classification maps and accuracy assessment matrices using standardised rubrics
- Field simulation exercises using pre-collected ground truth datasets to design and execute validation campaigns for land cover classifications
- Google Earth Engine coding labs with guided scripts and open exploration time for participants to experiment with large-scale time-series datasets
- Capstone project in which each participant or small team applies the full analytical pipeline to a dataset and challenge of their choosing, culminating in a professional briefing document and map product
Organisational Benefits
- Organisations gain staff capable of designing sophisticated remote sensing workflows without dependence on external consultants
- Enhanced ability to monitor agricultural productivity, land degradation, and ecosystem health using satellite data
- Improved accuracy and reliability of environmental assessments submitted to regulators and funding bodies
- Faster and more cost-effective land cover mapping and change detection through automation and cloud-based platforms
- Strengthened institutional capacity for early warning, disaster risk monitoring, and climate adaptation planning
- Better integration of earth observation data into evidence-based policy and programme decision-making
- Increased competitiveness in bidding for donor-funded projects requiring advanced geospatial technical expertise
- Reduced data procurement and processing costs through effective use of open-source imagery and tools
- A cadre of in-house analysts able to mentor junior GIS staff and build organisational geospatial literacy
Where Is Advanced Remote Sensing Data Analysis Training Offered?
This course runs in-person across: Dubai, Abuja, Nakuru, Nairobi, Zanzibar, Pretoria, Arusha, Johannesburg, Naivasha, Kampala, Accra, Kisumu, Addis Ababa, Dar es Salaam, Mombasa, Cape Town, Riyadh, Kigali, Lagos, plus Live Virtual (online) cohorts. Browse live dates for your city below.
Upcoming Schedules
| Location | Format | Dates | Price |
|---|
| Live Virtual |
Online |
5 Oct 2026 – 16 Oct 2026 |
KES 160,000 |
| Dubai, United Arab Emirates |
In-Person |
5 Oct 2026 – 16 Oct 2026 |
USD 8,200 |
| Abuja, Nigeria |
In-Person |
5 Oct 2026 – 16 Oct 2026 |
USD 5,600 |
| Nakuru, Kenya |
In-Person |
5 Oct 2026 – 16 Oct 2026 |
KES 230,000 |
| Nairobi, Kenya |
In-Person |
5 Oct 2026 – 16 Oct 2026 |
KES 230,000 |
| Zanzibar, Tanzania |
In-Person |
5 Oct 2026 – 16 Oct 2026 |
USD 5,000 |
| Pretoria, South Africa |
In-Person |
19 Oct 2026 – 28 Oct 2026 |
USD 6,600 |
| Arusha, Tanzania |
In-Person |
19 Oct 2026 – 28 Oct 2026 |
USD 4,200 |
| Johannesburg, South Africa |
In-Person |
19 Oct 2026 – 30 Oct 2026 |
USD 7,000 |
| Naivasha, Kenya |
In-Person |
19 Oct 2026 – 30 Oct 2026 |
KES 230,000 |
| Kampala, Uganda |
In-Person |
19 Oct 2026 – 30 Oct 2026 |
USD 3,800 |
| Accra, Ghana |
In-Person |
19 Oct 2026 – 30 Oct 2026 |
USD 7,600 |
| Kisumu, Kenya |
In-Person |
19 Oct 2026 – 28 Oct 2026 |
KES 240,000 |
| Addis Ababa, Ethiopia |
In-Person |
19 Oct 2026 – 28 Oct 2026 |
USD 4,800 |
| Dar es Salaam, Tanzania |
In-Person |
19 Oct 2026 – 30 Oct 2026 |
USD 4,600 |
+232 more dates — view live schedule →
Frequently Asked Questions
Do I need programming experience to attend this course?
Basic familiarity with any scripting language is helpful, but not a strict prerequisite. The Python sessions are structured progressively, starting from environment setup and data loading before advancing to classification and automation tasks. Participants with no coding background should expect to invest additional time during evening self-study sessions.
What software and hardware will I need to bring?
Participants must bring a personal laptop running Windows 10/11 or macOS with at least 8 GB RAM and 50 GB of free disk space. Please install QGIS 3.x and Anaconda (Python 3.9+) before arrival — installation guides will be provided upon registration. ENVI and SNAP are also used and will be available on training room workstations.
Will I receive a certificate upon completion?
Yes. Participants who attend at least 80% of sessions and complete the capstone project will receive a TrainingComplex Certificate of Completion in Advanced Remote Sensing Data Analysis, which can be used for professional development records and CPD portfolios.
What kinds of datasets will we work with during the training?
The course uses real-world datasets including Sentinel-2 multispectral imagery, Landsat-8/9 time series, ALOS PALSAR SAR scenes, and publicly available LiDAR point clouds from forestry and urban environments. Case study areas span East Africa, West Africa, and the Middle East to ensure regional relevance for most participants.
Is this course suitable for someone who completed a basic remote sensing course recently?
Yes, this programme is ideal for professionals who have completed an introductory remote sensing or GIS course and are ready to advance their technical skills. Participants should be comfortable with concepts such as band composites, basic image classification, and coordinate systems before attending.
