I'm a Copernicus Master in Digital Earth student specializing in Geospatial Data Science, with extensive experience in satellite image processing, Earth observation data analysis, and environmental monitoring. My expertise lies at the intersection of cutting-edge remote sensing technologies and advanced data science methodologies.
- π‘ Advanced Earth Observation: Multi-spectral and hyperspectral satellite data analysis
- π± Environmental Monitoring: Land use/land cover change detection and ecosystem analysis
- π¬ Deep Learning for EO: Neural networks for satellite image classification and segmentation
- π Water Resources: Global surface water mapping and flood monitoring using GEE
- π Geospatial Analytics: Large-scale environmental data processing and visualization
π₯ Competition Winner | π Scholarship Recipient | π€ AI/ML Expert | π°οΈ Earth Observation Specialist
π ML4EARTH Competition Winner β’ π«π· Eiffel Excellence Scholarship Winner β’ πΊοΈ COVID-19 Mapathon Winner β’ π Erasmus Mundus Student β’ π» Full-Stack GIS Developer
οΏ½ ML4EARTH: Foundation Models for EO - Competition-winning Earth observation foundation model implementation
π― Project Overview:
Award-winning implementation of foundation models for Earth observation data, demonstrating cutting-edge machine learning techniques for satellite image analysis.
β¨ Key Features:
- π Competition Winner: Top-performing solution in ML4EARTH challenge
- π€ Foundation Models: Advanced architectures for EO data
- οΏ½οΈ Multi-spectral Analysis: Comprehensive satellite data processing
π οΈ Tech Stack: Python PyTorch Transformers Satellite Data Deep Learning
οΏ½ U-Net Landsat 10-Class Classification - Deep learning for land cover classification using Landsat imagery
π― Project Overview:
Advanced U-Net implementation for pixel-level land cover classification using Landsat satellite imagery, achieving high-accuracy multi-class segmentation.
β¨ Key Features:
- π§ U-Net Architecture: Deep convolutional neural network for semantic segmentation
- οΏ½οΈ Landsat Integration: Optimized for Landsat multispectral bands
- ποΈ 10-Class Classification: Comprehensive land cover categories
- π High Accuracy: Optimized model performance and validation
- οΏ½ End-to-end Pipeline: Data preprocessing to model deployment
π οΈ Tech Stack: Python TensorFlow/Keras Landsat U-Net Computer Vision
π Land Cover Classification using GEE - Google Earth Engine-based large-scale land cover mapping
π― Project Overview:
Comprehensive land cover classification system leveraging Google Earth Engine's cloud computing platform for large-scale environmental monitoring and analysis.
β¨ Key Features:
- οΏ½ Google Earth Engine: Cloud-based geospatial analysis platform
- οΏ½ Multi-satellite Data: Sentinel-2, Landsat integration
- πΊοΈ Large-scale Mapping: Regional to global land cover classification
- π Temporal Analysis: Time-series land cover change detection
- οΏ½ Automated Workflows: Scalable processing pipelines
- π Accuracy Assessment: Robust validation methodologies
οΏ½οΈ Tech Stack: JavaScript Google Earth Engine Sentinel-2 Landsat Remote Sensing
|
|
|
|
|
|
|
|
|
|
|
|
|
AI & Machine Learning β’ Remote Sensing Consultancy β’ Geospatial Data Science β’ Earth Observation Research β’ Environmental Monitoring β’ GIS Development β’ Satellite Image Processing
Currently pursuing advanced research in Digital Earth technologies at the intersection of remote sensing, machine learning, and environmental science.
