![](/upload_mm/0/c/0/98c71e6d-2336-4a6a-9e68-00d1702c60fb_masiliunas2022_1_730485da_750x400.png)
Thesis subject
MSc thsis topic: AI for land cover change monitoring
Monitoring land cover change is essential for understanding and adapting to many challenges we are facing today from climate change to sustainable management of Earth’s resources. Automatic monitoring of changes on Earth’s surface allows rapid identification of hotspots of land transitions such as urbanization, deforestation, and cropland expansion, which allows law enforcement and policymakers to act upon.
Thanks to the improvements in Remote Sensing technologies, land cover change monitoring can be done at high spatial and temporal details. However, challenges still remain in identifying land cover changes globally, considering the regional differences in remotely sensed reflectance in certain land cover types, such as cropland, and the variations in the extent of transitions, e.g., large- vs small-scale forest plantations.
Advancements in machine learning algorithms can further enhance the quality of such monitoring. Therefore, this study aims to evaluate to what extent deep learning-based change monitoring methods can accurately map land cover changes and specific transitions (e.g., cropland expansion, urbanization, and deforestation). Deep learning architectures such as recurrent neural networks (RNN) and attention models (e.g., Transformer) can be investigated.
Objectives
- Develop a deep learning method for identifying a general land cover change in recent years
- Assess the performance of the method for different land cover change transitions
Literature
- Masolele, R.N., De Sy, V., Herold, M., Marcos, D., Verbesselt, J., Gieseke, F., Mullissa, A.G., & Martius, C. (2021). Spatial and temporal deep learning methods for deriving land-use following deforestation: A pan-tropical case study using Landsat time series. Remote Sensing of Environment, 264, 112600
- L. Mou and X. X. Zhu, "A Recurrent Convolutional Neural Network for Land Cover Change Detection in Multispectral Images", IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, 2018, pp. 4363-4366.
- Sefrin O, Riese FM, Keller S. Deep Learning for Land Cover Change Detection. Remote Sensing. 2021; 13(1):78.
Requirements
- Affinity work with large scale remote sensing datasets
Theme(s): Modelling & visualisation; Integrated Land Monitoring