He is currently a Postdoctoral Researcher at the The Institute for Analytical Sociology (IAS) at Linköping University. His work focuses on deep learning for poverty prediction, settlement mapping, and uncertainty quantification using multispectral and radar satellite imagery. His broader research vision involves creating trustworthy AI-EO systems that support global development goals (SDGs), improve decision-making, and provide robust, interpretable insights from large-scale satellite data.
Mohammad Kakooei
Postdoc
My research interests lie in Machine Learning (ML), Artificial Intelligence (AI), and Earth Observation (EO), with extensive experience in using satellite imagery and data-driven modeling to address global environmental and socio-economic challenges.
About me
Dr. Mohammad Kakooei is a researcher in Machine Learning (ML), Artificial Intelligence (AI), and Earth Observation (EO) with extensive experience in using satellite imagery and data-driven modeling to address global socio-economic and environmental challenges.
Brief facts
[2025 – Present ]
Researcher, Linköping University, The Institute for Analytical Sociology (IAS), Gothenburg, Sweden.
[2026 – Present ]
Senior Lecturer, Karlstad University, Geomatics, Department of Environmental and Life Sciences,, Sweden.
[2021 – 2025 ]
Postdoc Researcher, Chalmers University of Technology, Data science and Artificial Intelligence Division, Gothenburg, Sweden.
[2020 – 2021 ]
Postdoc Researcher, Babol Noshirvani University of Technology, Babol, Iran.
[2017 – 2018 ]
Visiting PHD Student, KTH (Royal Institute of Technology), Geoinformatics Division, Stockholm, Sweden.
Education
[2014 – 2020 ]
Ph.D., Electronics in Babol Noshirvani University of Technology, Babol, Iran.
Thesis title: Building Damage Assessment after Natural Disasters by Fusion of Earth Observationimages.
[2011 – 2014 ]
M.Sc., Electronics in Iran University of Science and Technology, Tehran, Iran.
Thesis title: Proposing Parallel Data Stream Clustering Algorithm Based on GPU.
[2006 – 2011 ]
B.Sc., Electronics in Shahid Beheshti University, Tehran, Iran.
Publications
2024
Mohammad Kakooei, Adel Daoud
(2024)
IEEE Transactions on Geoscience and Remote Sensing , Vol.62
Markus B. Pettersson, Mohammad Kakooei, Julia Ortheden, Fredrik D. Johansson, Adel Daoud
(2023)
PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023 , s.6165-6173
Research
Earth Observation & Remote Sensing
Earth Observation & Remote Sensing
- Large-scale mapping of urban and rural areas using multispectral (Landsat/Sentinel) and radar (Sentinel-1) imagery.
- Development of deep learning pipelines for:
- Fusion of EO imagery for disaster damage assessment.
- Time-series analysis of EO data, uncertainty quantification, and deep time-series models.
- Scalable data processing using Google Earth Engine, cloud computing, and GPU-based acceleration.
- Developing automated workflows for:
Computer Vision & Machine Learning
- Development of ML models for:
- Poverty estimation
- Urban structure analysis
- Wetland and crop mapping
- Land cover change detection
- Expertise in designing and training:
- Convolutional Neural Networks (CNNs)
- Self-supervised learning for EO
- Bayesian and probabilistic models
- Wavelet-based and time-series ML models
- Experience in:
- Semantic labeling of VHR images
- Spectral unmixing
- Random Forest, SVMs, K-means
- Neural network鈥揵ased segmentation
- GPU programming and high-performance computing using CUDA for scalable ML.
GIS & Geospatial Data Science
- Land use and land cover mapping at continental and national scales.
- Statistical spatial analysis of:
- Urban environmental variables
- Building height distributions
- Soil characteristics
- Creation of big EO datasets, integration of multiple data streams, and geo-big-data management.
- GIS workflows for policy-relevant applications (e.g., environmental monitoring, socio-economic datasets).
Teaching
Employment History