A stunning autumn ride through the forest. © Matthias Weigand - CC BY-SA 4.0
I love Geodata – I love Free and Open Source Software – I love cycling!
I am Matthias Weigand, a geographer and geoinformation scientist from Germany. I finished my PhD thesis entitled "Remote Sensing and Machine Learning for Detecting Urban Green – An Exemplary Analysis on the Distributive Justice in Germany" (german original "Fernerkundung und maschinelles Lernen zur Erfassung von urbanem Grün – eine Analyse am Beispiel der Verteilungsgerechtigkeit in Deutschland") at the Julius-Maximilians-University Würzburg in 2023. Since 2020 I am working as a researcher combining applied remote sensing, geoinformatics, data science and machine learning for applied geographic analyses. My main research focus is to use remote sensing image data in combination with diverse geographic data sets to gain new insights into the human habitat and its properties.
I especially enjoy tinkering with free and open source data to achieve all kinds of things for work as well as for private projects. When not nerding around, I spend my free time on my bike enjoying some gravel-ish rides.
Read more about me in my bio.
Research
My current research focus is on the use of large scale earth observation data. Here I will link my Publications.
Teaching
Between 2018 and 2020 I was teaching at University Augsburg. The block-course "Advanced Geodata Analysis and Machine Learning in R" is usually held in the summer semester.
Coding
To make my life easier, I tend to use overcomplicated shell scripts 📜. Occasionally, I will blog about them if they might be of any interest to others. All my dotfiles are available via my dotfiles repo.
To make working with large amounts of geotagged satellite images more convenienct, I stiched together a tool to convert large amounts of images into HDF5 files keeping the geographic reference intact: UKIS_sat2h5. Thus, the images are way more convenient to use in a Deep Learning data pipeline.
Also, I was annoyed by having to format markdown tables in my .Rmd files manually, so I created
and maintain the beautifyR R package.