The spatial analysis module helped me to build up my expertise in Geographic Information Systems (GIS).  I gained fundamental skills in geoinformatics where I learnt how to extract geospatial information to make spatial relations more explicit and operational.  We learned about map algebra, distance metrics, spatial query operators, interpolation methods (IDW, kriging, trend surface), Dijkstra network, vector and raster overlay, cost surface modeling, terrain catchment modelling (DTM, DSM), visual impact analysis, solar radiation analysis and surface run-off analysis.

In the spatial analysis and modelling module, I learnt in-depth spatial analysis techniques in remote sensing application and image analysis.  A wide range of topics were covered from image acquisition and image retrieval, image calibration and correction, DEM & DSM generation, mathematical principles for applying image filters (high pass, low pass, edge detectors), band ratios calculation, image segmentation, knowledge-based mage classification (KNN, Support Vector Machine, Random Forest), Convolution Neural Networks, semantic segmentation, cloud processing and image data cubes.

The Spatial Statistics module helped me acquire an understanding of the key methods and procedures in statistics and geostatistics of spatial data.  I learn how to communicate findings in a comprehensive and analytical manner.  The module provided a fundamental statistical principles of how to describe and perform inferential statistics.  I learn about point pattern analysis, spatial autocorrelation, cluster and dispersion, deterministic and probabilistic approaches of interpolation (Kriging, cross validation, variography and geographically weighted regression.  I learnt how to perform statistical analysis using R programming and ArcGIS Pro software.  

The advanced remote sensing module involved practical classes to master 'real world' application scenarios especially in the Copernicus context that enhanced our technical skills to perform advanced acquisition techniques (VHR multispectral, hyperspectral, Radar, Lidar, UAV data).  We learnt about image pre-processing (calibration, radiometric, correction, filtering, segmentation), image understanding and analysis (OBIA, CNN, advanced classifiers, class modelling and knowledge based representation).

​

This module gives us the fundamental understanding of the principles of active and multitemporal remote sensing and a broad knowledge of the different sensors available to process satellite and airborne data.  I learn how to perform data analysis to address specific methodological tasks using earth observation datasets.  I leant more about LiDAR (DEM, 3D points clouds, Synthetic Aperture Radar, Unmanned Aerial Vehicle,  data processing and analysis performed in dedicated softwares such as eMotion, cloudCompare, SNAP, Agisoft Photoscan and QGIS.  I further discovered how to operate a UAV flight (drone) simulation plan.    

The Artificial Intelligence module provided fundamental knowledge on machine learning and deep learning techniques.  I learn about the principles of supervised and unsupervised learning along with other machine learning paradigms (Random Forest, KNN, decision trees).  We were introduced the principles of classification and regression models.  I further learn about the principles of neural networks using convolution neural networks where we participated in the IEEE Data fusion contest 2021 to use our knowledge gained to address the multitemporal semantic change detection  MSD track challenge where we solve land cover change detection using classification tasks.

​

I learnt the principles of making good maps and charts and how to address visual communication of spatial issues.  The different topics we studied involves thematic mapping, web mapping and new methods of the visual geospatial displays to explore data.  We learnt hands-on practical tools in the exploration to solve spatial data and non-spatial data using Tableau, Leaflet, ArcGIS Pro and so on. We learn several methods of developing maps to be represented more qualitatively and quantitatively.    

This module focus in geo-application development using Object-Based Image Analysis (OBIA) concepts.  I learnt how to use Cognition Network Language (CNL) in the eCognition OBIA software environment to develop my own GUI with remote sensing application.  I further dig-in the OBIA concepts and theory to have a deeper understanding using CNL software skills to develop an eCognition Architect solution in the final project.  The CNL is a procedural computer language implemented in eCognition to enable automated and semi-automated implementation of complex, context dependent image analysis tasks.  

I learn about the state of the art spatial data infrastructure (SDI) technology.  This module covers a broad aspect of organizing geospatial data and information.  The conceptual strategies, importance of standardizing data models, metadata, spatial data sharing policies including intellectual property rights, security issues and open government data initiatives were broadly discussed.  I also discovered the Service Oriented Architecture (SOA), IT standards (OGC, ISO) to apply to practical research projects.  I also learn how to design and implement geodata models, publishing on geoprocessing services over the web.   

​