セクションアウトライン

  • Introduction

    すべてを折りたたむ すべてを展開する

    This course brings together two hands‑on, practice‑oriented tutorials: Field data collection with Mergin Maps and Remote Sensing Image Classification with the Semi-Automatic Classification plugin. These tutorials help you build practical skills in modern, open‑source geospatial workflows. The two components are intentionally connected: supervised image classification depends on high‑quality training data, and this course shows you how to collect that data in the field using Mergin Maps before using it to train your classification model in QGIS.

    Whether you are new to field data collection or taking your first steps in land‑cover mapping, the course guides you from preparing your survey, to gathering reliable ground‑truth points, to producing a validated land‑cover map with confidence.

    Learning Objectives

    After completing this course, you will be able to:

    • Collect, synchronise, and manage field data using Mergin Maps.
    • Apply AI‑assisted segmentation and semi‑automatic workflows for preparing a ground truth dataset.
    • Prepare satellite imagery and perform supervised classification in QGIS.
    • Understand the full pipeline from raw imagery to a validated classification map.

     

    Software Requirements

    All exercises are designed for QGIS 3.44 LTR. Using this version ensures compatibility with the plugins and workflows demonstrated in the tutorials.

     

    Acknowledgements

    This course was developed by Hans van der Kwast for IHE Delft Institute for Water Education, QWAST, and the Geo‑ICT Training Center. We gratefully acknowledge TerraLab for developing the AI Segmentation plugin and for providing an alternative one‑click installer for the Semi‑Automatic Classification Plugin (SCP), which greatly improves accessibility for learners.

  • Field Data Collection

  • Remote Sensing Image Classification