Aim and content

This PhD course, “From Point to Pixel,” is designed to equip students with the essential tools and skills needed to transform ground truth sample data into high-resolution pixel-level estimates, which can then be aggregated into comprehensive maps. The primary goal of the course is to provide students with the expertise necessary to quantify the spatial landscape accurately. These maps serve as crucial resources for making informed decisions related to, for example, biodiversity conservation or facilitating the transition to a greener society through sustainable forest management.

Key Course Objectives:

Ground Truth Data Handling: Students will learn how to develop a sampling strategy, collect, organize, and preprocess ground truth observations efficiently. This includes data cleaning, quality control, and geospatial data handling techniques.

Advanced Modeling Techniques:

The course will delve into advanced modeling methods that allow students to establish robust relationships between ground truth observations and remotely sensed variables. This includes statistical modeling, machine learning, and geospatial modeling approaches.

Pixel-Level Estimation:

Students will be introduced to relevant RS data to be used in combination with ground truthing and machine learning to generate pixel-level estimates, enabling them to produce high-resolution maps that accurately represent the environmental parameters under study.

Spatial Landscape Mapping:

Students will learn how to aggregate pixel-level estimates into comprehensive maps, providing a detailed quantification of the spatial landscape. These maps can be used for various applications, such as assessing habitat quality for biodiversity and estimating available wood resources for a sustainable green transition.

By the end of “From Point to Pixel,” participants will possess a powerful toolbox of techniques and methodologies to produce accurate, high-resolution maps that aid in biodiversity conservation, sustainable wood resource management, and other critical aspects of environmental science. This course equips students with the skills needed to make data-driven decisions and contribute to the development of a more sustainable and environmentally conscious society.

Learning outcome

Knowledge:

• … of remotely sensed datatypes and their strength and weaknesses.
• … of ground truth sampling designs and methodologies.
• … of methods for statistical modeling, machine learning, and geospatial modeling approaches.

 Skills:

• To develop sampling strategies.
• To collect, organize, and preprocess ground truth observations efficiently.
• To generate maps from pixel-based remote sensing.
• To make data-driven decisions for a sustainable society.

Competences:

• Ground Truth Data Handling: Students will gain proficiency in developing effective sampling strategies, collecting, organizing, and preprocessing ground truth observations efficiently.
• Participants will acquire expertise in employing advanced modeling methods to establish robust relationships between ground truth observations and remotely sensed variables.
• The course will equip students with the knowledge and skills necessary to utilize remote sensing data in conjunction with ground truthing and machine learning techniques to generate precise pixel-level estimates.
• Participants will learn how to aggregate pixel-level estimates into comprehensive maps, enabling detailed quantification of the spatial landscape. This involves understanding spatial patterns and processes, as well as techniques for synthesizing and visualizing complex geospatial information.

Teaching and learning methods

A variety of teaching and learning methods are applied to reach the goals of the course:

Lectures:

We use lectures to provide students with tools necessary to engage in the practical exercises and project work.

Practical exercises:

Throughout the course, students will work on real-world case studies and projects to apply their knowledge and skills to solve environmental challenges.

Project work:

We encourage students to bring a case study that they wish to work with during the course.

Type af assessment:

 Written assignment on own project or project handed out by the course responsible must be completed and approved by the course responsible.

There will be a course fee of DKK 700. The fee covers a course dinner during the lecture week and snacks, fruit, and coffee/tee served during course days. Lunch will be self-organized.

Disclaimer:
DDSA has explicit permission from Arcanic and the owners of the https://phdcourses.dk/ website to display the courses on ddsa.dk.