Detecting Log-yards in Europe with big data and GeoAI
- 2nd supervisor is open for discussion. Links with Wageningen University
- Suggested elective courses: anything related to Spatial Statistics, machine learning and advanced remote sensing (including OBIA)
To analyze wood flows and linking the wood industry to forest resources it is useful to know all the saw mills, panel mills, bioenergy facilities, and pulp mills in Europe. A large database with some 4000 mills and facilities (Bozzolan et al 2024) exists, but this database is not complete. This database can be potentially improved bu detecting the so called log-yards with a combination of remote sensing and GeoAI techniques.
This project will try to detect all log yards in Europe (or a pilot with one country), and validate this with the existing dataset. The picture on top of this topic is a good example, log yards have a distinct pattern, texture, spectral signature and size. The size of the log yard is also a good indicator of the mills capacity.
The topic will study how well log-yards can be discriminated from other similar looking spatial features.
- Bozzolan, N., Grassi, G., Mohren, F., & Nabuurs, G.-J. (2023). Options to improve the carbon balance of the harvested wood products sector in four EU countries. GCB Bioenergy, 16, e13104. https://doi.org/10.1111/gcbb.13104