Advanced Remote Sensing and Earth Observation

Start course

Duration

Content

Image Classification Principles

Remote Sensing Image Classification

Key Concepts

The Classification Process

Image Classification: Practical

Q&A

Essay Format Questions

Glossary of Key Terms

Conclusion: Classification Methodology

Basic Literature

The course is designed for vocational training in information extraction from satellite data, with a focus on image classification and segmentation. It covers the fundamental concepts in remote sensing image classification, including the concepts of image and feature space, supervised and unsupervised approaches, various classification algorithms like maximum likelihood, and the crucial step of validating results using error or confusion matrices.

Image classification is utilised in various regional projects. The European Commission requires national governments to verify farmers' claims regarding crop subsidies. To fulfil these requirements, national governments hire companies to conduct an initial assessment using image classification techniques, which is subsequently followed by field checks.

This course also covers key concepts, methods, and challenges in remote sensing image classification.

Special attention is also given to free, open-source systems for viewing and interpreting satellite imagery. It enables participants to efficiently collect up-to-date information about their projects and monitor changes over time.

Syllabus

Access to a growing archive of satellite imagery and platforms (4 hours lecture)

Critical datasets selection for experimentation (4 hours exercises)

The availability of new tools to facilitate image processing and interpretation, including improved EO data selection, visualisation, review, and analysis functions

Processing EO Satellite imagery including cloud masking and image enhancement (2 hours lecture and 4 hours exercises)

Introduction to landcover maps (2 hours lecture)

Remote sensing indices (3 hours lecture and 3 hours exercises)

Image Classification (9 hours lecture and 9 hours exercises)

methods
Feature extraction
clustering
training set selection
supervised and unsupervised image classification
landcover mapping

The homework will include: 20 hours literature analysis, RS terminology, and individual study.

Objectives and Competences

In this course, students will learn about advanced information extraction from satellite data (Focus: Analysis of Satellite Images).

The main objectives are:

To provide the knowledge, skills, and understanding necessary for advanced information extraction from satellite data.
To develop an understanding of the fundamentals of Remote Sensing image classification.

Intended Learning Outcomes

Students who successfully complete the course will be able to:

Select band/feature.
Conduct practical exercises in image classification using available data and software.
Generate independent, robust, and consistent reporting needs for image classification and estimate pista orchards/tree areas and their changes over time at regional, farm, and plot levels.
Collaborate with Pista growers to do knowledge-based classification.