ML4EO 2024

Keynote Speakers

Oral Presentations

Mapping human impacts on rangelands with Earth observation and quantile regression forests 

Guy Lomax (University of Exeter)* 

Multi-spatial-temporal remote sensing and machine learning for mapping land use on peatlands in Ireland 

Wahaj WH Habib (Trinity College Dublin)*; John Connolly (Trinity College Dublin) 

The beat of the savanna: Towards non-invasive vibrational monitoring of sub-Saharan wildlife

Tarje Nissen-Meyer (University of Exeter); Rene Steinmann (GFZ Potsdam); Beth Mortimer (University of Oxford) 

Bridging the Data Gap: combining Synthetic and Satellite Data for Agricultural Land Monitoring 

Enes Hisam (GMV NSL); Rossana Gini (GMV NSL)*; David Miraut Andrés (GMV); Raúl Rodríguez Juárez (GMV); Marta Toro Bermejo (GMV); David de la Fuente Blanco (GMV); Jesús Gimeno (Universidad de Valencia); Marcos Fernández Marín (Universidad de Valencia) 

Can we use satellite imagery and artificial intelligence to automate habitat monitoring in key biodiversity areas?  

Andy Wallace (ATOS) 

Exploring Sentinel-1-based Change Detection Approaches for Updating the Antarctic Coastline 

Aliaksandra S. Skachkova (British Antarctic Survey)* 

Predicting Zostera marina seagrass carbon stock using machine learning and earth observation to support carbon accounting 

Nicola Wilson (University of Exeter)*

Prediction of diagnostic pigments in the ocean by machine-learning  

Angus Laurenson (Plymouth Marine Laboratory)* 

Evaluation of accuracy and uncertainty for satellite-derived coastal bathymetry prediction using Machine Learning and Hybrid Spatial models 

Paul Harris (Rothamsted Research)*; Conor Cahalane (Maynooth University); Xavier Monteys (GSI); Gema Casal (Maynooth University) 

Reconstructing the spatiotemporal evolution of the global interior ocean’s anthropogenic carbon sink using deep learning 

Tobias F Ehmen (University of Exeter)*; Neill Mackay (University of Exeter); Andrew Watson (University of Exeter) 

Pivoting Perspectives: Adapting Medical Artificial Intelligence Techniques for Environmental Sample Analysis 

Claire Barnes (Swansea University)*; Eloise Smith (Swansea University) 

Using AI with satellite-borne cloud radar data  

Cyril J Morcrette (Met Office, UK)* 

Object Detection on Satellite Imaging for Sustainable Water Harvesting Placements in Maasai Region, Tanzania 

Roshan Taneja*; Yuvraj Taneja 

Improving the Proactive and Reactive Maintenance of Water Pipe Networks using InSAR and Machine Learning 

Andrew Watson (SatSense)*; Sarah Douglas (SatSense); Tom Ingleby (SatSense) 

Detection of landslides from space using open-source data.  

Kathryn Leeming (BGS) 

DisasterScope: A Dataset and RTMDet-based Methodology for Object Detection in Disaster-Related Remote Sensing Images 

Zhipeng Liu (University of Exeter)*

Machine Learning and historical impact data for predicting Tropical Cyclone impacts  

Elizabeth Galloway (University of Exeter) 

Rapid Identification of Flood-Affected Areas Using a Sequential Ensemble Algorithm 

RUI GUO (Beijing Normal University) 

Exploring the potential of freely available remote sensing data and machine learning to monitor above ground tree biomass in the Miombo woodlands of Zambia 

Kennedy Kanja (Lancaster University)* 

Using an ensemble of high-dimensionality convolutional neural networks to map continent scale forest cover and changes over time. 

James Brown (Geospatial Intelligence Ltd)*; Rob Coorey (Geospatial Intelligence Pty Ltd); Jeremy Repetto (Geospatial Intelligence Pty Ltd); Sam Lewis (Geospatial Intelligence Pty Ltd) 

Changes in forest coverage in Mexico and their relationship with soil moisture and organic carbon content 

Luis Felipe Castelblanco Rivera (Universidad Nacional Autónoma de México)*; Mario Antonio Guevara Sanatamaría (Universidad Nacional Autónoma de México) 

Mapping the Spatio-Temporal Distribution of Burned Areas in the Amazon from 2001 to 2020: An Ensemble Modelling Approach 

Oscar Rodriguez de Rivera Ortega (University of Exeter)*; Paula Moraga (KAUST); Jonatan Gonzalez (Universidad Miguel Hernandez); Mohamed Abid (KAUST) 

Tree genera classifications using Sentinel-2 time-series extracted from PlotToSat 

Milto Miltiadou (University of Cambridge)*; Stuart Grieve ( Queen Mary University of London); Paloma Ruiz-Benito (Universidad de Alcalá); Julen Astigarraga (Universidad de Alcalá); Verónica Cruz-Alonso (Universidad de Alcalá); Emily Lines (Department of Geography, University of Cambridge) 

Building trust in Machine Learning for commercial forest carbon applications. 

Hugh Graham (Permian Global, University of Exeter)*, Andy Cunliffe (University of Exeter), Chris Philipson (Permian Global) 

Land cover classification in the UK using geospatial foundation models 

Remy Vandaele (University of Exeter)*; Edward Pope (Met Office); Hywel T P Williams (University of Exeter) 

Insight4EO: Onboard Intelligence for Advanced Decision-Making and EO Data Processing 

Michael Wilby (Deimos Space UK)*; Juan Ignacio Bravo Pérez-Villar (Deimos Engineering and Systems); Álvaro Morón Elorza (Deimos Engineering and Systems); Enrique Sepúlveda Jorcano (Deimos Engineering and Systems); Robert Hinz (Deimos Engineering and Systems); Rohaan Ahmed (Deimos Space UK) 

Application of RandLANet to the semantic segmentation of low-density photogrammetric 3D point clouds 

Lucy Main (Ordnance Survey)* 

GlobeNet: Geo-Foundation model development 

Steve Coupland (Ordnance Survey)* 

Context-Aware Satellite Image Change Detection via Multi-Scale Transformer Model 

Zeinab Gharib Bafghi (Osnabrück University)*; Peter Reinartz (German Aerospace Center) 

The Earth Observation Data Life Cycle and its Environmental Impacts 

Markus Mueller (University of Exeter)* 

Enhancing Machine Learning Performance for Topography Modeling through Feature Tuning 

Joseph Paulo (University of Exeter)* 

The Need for Expert-Guided Deep Learning to Digitally Monitor Pollen in the Environment 

Ann Power (University of Exeter)*; Claire Barnes (Swansea University); Eloise Smith (Swansea University)

Training accuracy in building footprints detection for OpenStreetMap 

Anna Zanchetta (The Alan Turing Institute)*; Kshitij Sharma (Humanitarian OSM Team); Emran ALCHIKH ALNAJAR (Humanitarian OSM Team) 

Estimating vehicle flows in global cities for environmental applications using high-resolution satellite imagery and deep learning 

Annalisa Sheehan (St George’s, University London)* 

Is a Satellite Image Worth a Thousand Data Points? Comparing Machine Learning Approaches for Predicting Social and Environmental Inequalities Across England 

Antje Barbara Metzler (Alan Turing Institute)*; Dani Arribas-Bel (Alan Turing Institute); Martin Fleischmann (Charles University) 

Optimising biodiversity through machine learning 

Kamaran Fathulla (University of Aberdeen)*; Annett Frick (LUP – Luftbild Umwelt Planung GmbH); Nastasja Scholz (LUP – Luftbild Umwelt Planung GmbH) 

Leveraging Deep-Learning Approaches with Spatial Context for Enhanced Surface Solar Irradiance Estimation from Third-Generation Geostationary Satellite Imagery 

Vadim Becquet (Mines Paris PSL Centre OIE)*; Hadrien Verbois (Mines Paris PSL Centre OIE); Philippe Blanc (Mines Paris PSL Centre OIE); Yves-Marie Saint-Drenan (Mines Paris PSL Centre OIE) 

Learning-Based Predictive Scheduling for Multiple Agile Satellites with Task Arrivals During Mission 

Abhijit Chatterjee (University of Exeter)*; Ratnasingham Tharmarasa (McMaster University) 

Deep Learning for Improved Phase Unwrapping 

Eilish O’Grady (University of Leeds)*; Andy Hooper (University of Leeds); David C Hogg (University of Leeds); Matthew Gaddes (University of Leeds) 

Phase-Noise Suppression in InSAR Phase Using a Complex-Valued Image-Translation Deep Learning Model 

Gopal Singh Phartiyal (School of Earth and Environment, University of Leeds)*; Andy Hooper (School of Earth and Environment, University of Leeds); David C Hogg (University of Leeds); Eilish O’Grady (University of Leeds); Matthew Gaddes (University of Leeds); Milan Lazecky (University of Leeds) 

Posters

• Identifying deforestation drivers in Cameroon using deep learning and Earth observation data. Amandine Debus (University of Cambridge), Emilie Beauchamp (International Institute for Sustainable Development), Justin Kamga (Forêts et Dévelopmments Rural, FODER), Astrid Verhegghen (European Commission Joint Research Centre & ARHS Developments Italia S.R.L), Christiane Zébazé (Forêts et Dévelopmments Rural, FODER), Emily R. Lines (University of Cambridge) 
• Towards a framework for uncertainty propagation from satellite data to land cover maps and their downstream application. Anna Pustogvar (National Physical Laboratory, University of Leicester), Bernardo Mota (National Physical Laboratory), Samuel E. Hunt (National Physical Laboratory), Heiko Balzter (University of Leicester, National Centre for Earth Observation) 
• Living England: an earth-observation derived national scale habitat map. Becky Trippier (Natural England) 
• PeaSat: using satellite imagery to estimate yield of vining peas. Ben J Hockridge (ADAS) 
• Estimating Canopy Height in Tropical Forests: Integrating Airborne LiDAR and Multi-Spectral Optical Data with Machine Learning. Brianna J Pickstone (University of Exeter) 
• From pixel to peat: Mapping England’s Peatland. Craig C Dornan (NE) 
• Rates and Drivers of Cliff Erosion in England from over 20 Years of LiDAR Observations Erosion. Cristina Coker (University of Exeter) 
• Catastrophe modelling of health costs and asset losses due to future tsunamis over Sumatra and Java, Indonesia. Dimitra M Salmanidou (University College London) 
• Using continual pretraining with a geospatial foundation model. Geoffrey J Dawson (IBM), Chris Dearden (STFC), Andrew Taylor (STFC), Helen Tamura-Wicks (IBM Research), Paolo Fraccaro (IBM UK), Anne Jones (IBM Research) 
• A remote sensing foundation model for the British Isles. Helen Tamura-Wicks (IBM Research), Andrew Taylor (STFC), Chris Dearden (STFC), Geoffrey J Dawson (IBM), Anne Jones (IBM Research), Paolo Fraccaro (IBM UK) 
• Application of machine learning to forecast agricultural drought impacts for large scale sub-seasonal drought monitoring in Brazil. Joseph W Gallear (Rothamsted Research), Marcelo Valadares Galdos (Rothamsted Research), Marcelo Zeri (CEMADEN), Andrew Hartley (Met Office) 
• The Blue Belt Programme: Using Earth Observation Data to Support Global Marine Environmental Protection. Lewis Brady (Marine Management Organisation), Emma Harvey (Marine Management Organisation) 
• Machine Learning Volcanic Ash Detection. Måns Holmberg (Met Office), Cameron Saint (Met Office) 
• High Impact Weather in the Mid-Latitudes: A Neural Network Approach to Identifying Dry Intrusion Outflows. Owain L Harris (University of Exeter) 
• Potential of Renewable Distributed Energy Resources using Computer Vision and Earth Observation Data. Owen G.W. Saunders (University of Exeter), Cesar Angeles (University of Exeter) 
• Satellite detection of coal mine methane emissions using machine learning. Sarah Shannon (Ember), Sabina Assan (Ember) 
• Emulating melt ponds on sea ice with neural networks. Simon Driscoll (University of Reading) 
• Complex dynamical insights to air quality interplay in urban spaces: A case of cities co movements and comparison. Syed Shariq Husain (OP Jindal Global University) 
• A machine learning based climatology of dissolved organic carbon. Thelma Panaïotis (National Oceanography Centre), Jamie Wilson (University of Liverpool), BB Cael (National Oceanography Centre) 
• Hybrid ANN and Object-Based Image Analysis for Slum Detection and Monitoring in Abidjan and Karachi: An Operational Approach. Tomáš Bartaloš (GISAT), Jan Kolomazník (GISAT) 
• Mapping and quantifying the invasive species Prosopis in the rangelands of south-western Kalahari with multi-scale remote sensing. Glenn Slade, Andrew Cunliffe, Hugh Graham, Jeremy Perkins3, Karen Anderson (University of Exeter)