Girmaw Abebe Tadesse, Caleb Robinson, Charles Mwangi, Esther Maina, Joshua Nyakundi, Luana Marotti, Gilles Quentin Hacheme, Hamed Alemohammad, Rahul Dodhia, Juan M. Lavista Ferres
In 2023, 58.0% of the African population experienced moderate to severe food insecurity, with 21.6% facing severe food insecurity. Land-use and land-cover maps provide crucial insights for addressing food insecurity by improving agricultural efforts, including mapping and monitoring crop types and estimating yield. The development of global land-cover maps has been facilitated by the increasing availability of earth observation data and advancements in geospatial machine learning. However, these global maps exhibit lower accuracy and inconsistencies in Africa, partly due to the lack of representative training data. To address this issue, we propose a data-centric framework with a teacher-student model setup, which uses diverse data sources of satellite images and label examples to produce local land-cover maps. Our method trains a high-resolution teacher model on images with a resolution of 0.331 m/pixel and a low-resolution student model on publicly available images with a resolution of 10 m/pixel. The student model also utilizes the teacher model's output as its weak label examples through knowledge transfer. We evaluated our framework using Murang'a county in Kenya, renowned for its agricultural productivity, as a use case. Our local models achieved higher quality maps, with improvements of 0.14 in the F1 score and 0.21 in Intersection-over-Union, compared to the best global model. Our evaluation also revealed inconsistencies in existing global maps, with a maximum agreement rate of 0.30 among themselves. Our work provides valuable guidance to decision-makers for driving informed decisions to enhance food security.
Aditya Kulkarni, Tharun Mohandoss, Daniel Northrup, Ernest Mwebaze, Hamed Alemohammad
Semantic segmentation of satellite imagery is a common approach to identify patterns and detect changes around the planet. Most of the state-of-the-art semantic segmentation models are trained in a fully supervised way using Convolutional Neural Network (CNN). The generalization property of CNN is poor for satellite imagery because the data can be very diverse in terms of landscape types, image resolutions, and scarcity of labels for different geographies and seasons. Hence, the performance of CNN doesn't translate well to images from unseen regions or seasons. Inspired by Conditional Generative Adversarial Networks (CGAN) based approach of image-to-image translation for high-resolution satellite imagery, we propose a CGAN framework for land cover classification using medium-resolution Sentinel-2 imagery. We find that the CGAN model outperforms the CNN model of similar complexity by a significant margin on an unseen imbalanced test dataset.
Hamed Alemohammad, Kevin Booth
Regularly updated and accurate land cover maps are essential for monitoring 14 of the 17 Sustainable Development Goals. Multispectral satellite imagery provide high-quality and valuable information at global scale that can be used to develop land cover classification models. However, such a global application requires a geographically diverse training dataset. Here, we present LandCoverNet, a global training dataset for land cover classification based on Sentinel-2 observations at 10m spatial resolution. Land cover class labels are defined based on annual time-series of Sentinel-2, and verified by consensus among three human annotators.
Tharun Mohandoss, Aditya Kulkarni, Daniel Northrup, Ernest Mwebaze, Hamed Alemohammad
Multi-spectral satellite imagery provides valuable data at global scale for many environmental and socio-economic applications. Building supervised machine learning models based on these imagery, however, may require ground reference labels which are not available at global scale. Here, we propose a generative model to produce multi-resolution multi-spectral imagery based on Sentinel-2 data. The resulting synthetic images are indistinguishable from real ones by humans. This technique paves the road for future work to generate labeled synthetic imagery that can be used for data augmentation in data scarce regions and applications.
Yoni Nachmany, Hamed Alemohammad
Semantic segmentation of land cover classes is fundamental for agricultural and economic development work, from sustainable forestry to urban planning, yet existing training datasets have significant limitations. To generate an open and comprehensive training library of high resolution Earth imagery and high quality land cover classifications, public Sentinel-2 data at 10 m spatial resolution was matched with accurate GlobeLand30 labels from 2010, which were filtered by agreement with an intermediary Sentinel-2 classification at 20 m produced during atmospheric correction. Scene-level classifications were predicted by Random Forests trained on valid reflectance data and the filtered labels, and achieved over 80% model accuracy for a variety of locations. Further work is required to aggregate individual scene classifications for annual labels and to test the approach in more locations, before crowdsourcing human validation. The goal is to create a sustained community-wide effort to generate image labels not only for land cover, but also very specific images for major agriculture crops across the world and other thematic categories of interest to the global development community.
Alexandre Lacoste, Evan David Sherwin, Hannah Kerner, Hamed Alemohammad, Björn Lütjens, Jeremy Irvin, David Dao, Alex Chang, Mehmet Gunturkun, Alexandre Drouin, Pau Rodriguez, David Vazquez
Recent progress in self-supervision shows that pre-training large neural networks on vast amounts of unsupervised data can lead to impressive increases in generalisation for downstream tasks. Such models, recently coined as foundation models, have been transformational to the field of natural language processing. While similar models have also been trained on large corpuses of images, they are not well suited for remote sensing data. To stimulate the development of foundation models for Earth monitoring, we propose to develop a new benchmark comprised of a variety of downstream tasks related to climate change. We believe that this can lead to substantial improvements in many existing applications and facilitate the development of new applications. This proposal is also a call for collaboration with the aim of developing a better evaluation process to mitigate potential downsides of foundation models for Earth monitoring.
Seyed Hamed Alemohammad, Dara Entekhabi
Several passive microwave satellites orbit the Earth and measure rainfall. These measurements have the advantage of almost full global coverage when compared to surface rain gauges. However, these satellites have low temporal revisit and missing data over some regions. Image fusion is a useful technique to fill in the gaps of one image (one satellite measurement) using another one. The proposed algorithm uses an iterative fusion scheme to integrate information from two satellite measurements. The algorithm is implemented on two datasets for 7 years of half-hourly data. The results show significant improvements in rain detection and rain intensity in the merged measurements.
Reza Ardakanian, Seyed Hamed Alemohammad
Coastal regions have a high social, economical and environmental importance. Due to this importance the sea level fluctuations can have many bad consequences. In this research the correlation between the increasing trend of temperature in coastal stations due to Global Warming and the Caspian Sea level has been established. The Caspian Sea level data has been received from the Jason-1 satellite. It was resulted that the monthly correlation between the temperature and sea level is high and also positive and almost the same for all the stations. But the yearly correlation was negative. It means that the sea level has decreased by the increase in temperature.
Seyed Hamed Alemohammad, Reza Ardakanian, Akbar Karimi
Predicting future probable values of model parameters, is an essential pre-requisite for assessing model decision reliability in an uncertain environment. Scenario Analysis is a methodology for modelling uncertainty in water resources management modelling. Uncertainty if not considered appropriately in decision making will decrease reliability of decisions, especially in long-term planning. One of the challenges in Scenario Analysis is how scenarios are made. One of the most approved methods is statistical modelling based on Auto-Regressive models. Stream flow future scenarios in developed basins that human has made changes to the natural flow process could not be generated directly by ARMA modelling. In this case, making scenarios for monthly rainfall and using it in a water resources system model makes more sense. Rainfall is an ephemeral process which has zero values in some months which introduces some limitations in making use of monthly ARMA model. Therefore, a two stage modelling approach is adopted here which in the first stage yearly modelling is done. Within this yearly model three ranges are identified: Dry, Normal and Wet. In the normal range yearly ARMA modelling is used. Dry and Wet range are considered as random processes and are modeled by frequency analysis. Monthly distribution of rainfall, which is extracted from available data from a moving average are considered to be deterministic and fixed in time. Each rainfall scenario is composed of a yearly ARMA process super-imposed by dry and wet events according to the frequency analysis. This modelling framework is applied to available data from three rain-gauge stations in Iran. Results show this modelling approach has better consistency with observed data in comparison with making use of ARMA modelling alone.
Seyed Hamed Alemohammad, Alexandra G. Konings, Thomas Jagdhuber, Mahta Moghaddam, Dara Entekhabi
Understanding the scattering mechanisms from the ground surface in the presence of different vegetation densities is necessary for the interpretation of P-band Synthetic Aperture Radar (SAR) observations and for the design of geophysical retrieval algorithms. In this study, a quantitative analysis of vegetation and soil scattering mechanisms estimated from the observations of an airborne P-band SAR instrument across nine different biomes in North America is presented. The goal is to apply a hybrid (model- and eigen- based) three component decomposition approach to separate the contributions of surface, double-bounce and vegetation volume scattering across a wide range of biome conditions. The decomposition makes no prior assumptions about vegetation structure. We characterize the dynamics of the decomposition across different North American biomes and assess their characteristic range. Impacts of vegetation cover seasonality and soil surface roughness on the contributions of each scattering mechanism are also investigated. Observations used here are part of the NASA Airborne Microwave Observatory of Subcanopy and Subsurface (AirMOSS) mission and data have been collected between 2013 and 2015.
Naomi Simumba, Nils Lehmann, Paolo Fraccaro, Hamed Alemohammad, Geeth De Mel, Salman Khan, Manil Maskey, Nicolas Longepe, Xiao Xiang Zhu, Hannah Kerner, Juan Bernabe-Moreno, Alexandre Lacoste
Geospatial Foundation Models (GeoFMs) are transforming Earth Observation (EO), but evaluation lacks standardized protocols. GEO-Bench-2 addresses this with a comprehensive framework spanning classification, segmentation, regression, object detection, and instance segmentation across 19 permissively-licensed datasets. We introduce ''capability'' groups to rank models on datasets that share common characteristics (e.g., resolution, bands, temporality). This enables users to identify which models excel in each capability and determine which areas need improvement in future work. To support both fair comparison and methodological innovation, we define a prescriptive yet flexible evaluation protocol. This not only ensures consistency in benchmarking but also facilitates research into model adaptation strategies, a key and open challenge in advancing GeoFMs for downstream tasks. Our experiments show that no single model dominates across all tasks, confirming the specificity of the choices made during architecture design and pretraining. While models pretrained on natural images (ConvNext ImageNet, DINO V3) excel on high-resolution tasks, EO-specific models (TerraMind, Prithvi, and Clay) outperform them on multispectral applications such as agriculture and disaster response. These findings demonstrate that optimal model choice depends on task requirements, data modalities, and constraints. This shows that the goal of a single GeoFM model that performs well across all tasks remains open for future research. GEO-Bench-2 enables informed, reproducible GeoFM evaluation tailored to specific use cases. Code, data, and leaderboard for GEO-Bench-2 are publicly released under a permissive license.
Denys Godwin, Hanxi Li, Michael Cecil, Hamed Alemohammad
Filling cloudy pixels in multispectral satellite imagery is essential for accurate data analysis and downstream applications, especially for tasks which require time series data. To address this issue, we compare the performance of a foundational Vision Transformer (ViT) model with a baseline Conditional Generative Adversarial Network (CGAN) model for missing value imputation in time series of multispectral satellite imagery. We randomly mask time series of satellite images using real-world cloud masks and train each model to reconstruct the missing pixels. The ViT model is fine-tuned from a pretrained model, while the CGAN is trained from scratch. Using quantitative evaluation metrics such as structural similarity index and mean absolute error as well as qualitative visual analysis, we assess imputation accuracy and contextual preservation.
Yannis Kalantidis, Laura Sevilla-Lara, Ernest Mwebaze, Dina Machuve, Hamed Alemohammad, David Guerena
This is the proceedings of the Computer Vision for Agriculture (CV4A) Workshop that was held in conjunction with the International Conference on Learning Representations (ICLR) 2020. The Computer Vision for Agriculture (CV4A) 2020 workshop was scheduled to be held in Addis Ababa, Ethiopia, on April 26th, 2020. It was held virtually that same day due to the COVID-19 pandemic. The workshop was held in conjunction with the International Conference on Learning Representations (ICLR) 2020.
Daniela Szwarcman, Sujit Roy, Paolo Fraccaro, Þorsteinn Elí Gíslason, Benedikt Blumenstiel, Rinki Ghosal, Pedro Henrique de Oliveira, Joao Lucas de Sousa Almeida, Rocco Sedona, Yanghui Kang, Srija Chakraborty, Sizhe Wang, Carlos Gomes, Ankur Kumar, Myscon Truong, Denys Godwin, Hyunho Lee, Chia-Yu Hsu, Rohit Lal, Ata Akbari Asanjan, Besart Mujeci, Disha Shidham, Trevor Keenan, Paulo Arevalo, Wenwen Li, Hamed Alemohammad, Pontus Olofsson, Christopher Hain, Robert Kennedy, Bianca Zadrozny, David Bell, Gabriele Cavallaro, Campbell Watson, Manil Maskey, Rahul Ramachandran, Juan Bernabe Moreno
This paper presents Prithvi-EO-2.0, a new geospatial foundation model that offers significant improvements over its predecessor, Prithvi-EO-1.0. Trained on 4.2 million global time series samples from NASA's Harmonized Landsat and Sentinel-2 data archive at 30-m resolution, the new model incorporates temporal and location embeddings for enhanced performance across various geospatial tasks. Through extensive benchmarking with GEO-Bench, the model outperforms the previous Prithvi-EO model by 8% across a range of tasks. It also outperforms six other geospatial foundation models when benchmarked on remote sensing tasks from different domains and resolutions (i.e. from 0.1 m to 15 m). The results demonstrate the versatility of the model in both classical Earth observation and high-resolution applications. Early involvement of end-users and subject matter experts (SMEs) allowed constant feedback on model and dataset design, enabling customization across diverse SME-led applications in disaster response, land cover and crop mapping, and ecosystem dynamics monitoring. Prithvi-EO-2.0 is available as an open-source model on Hugging Face and IBM TerraTorch, with additional resources on GitHub. The project exemplifies the Trusted Open Science approach embraced by all involved organizations.
Johannes Jakubik, Sujit Roy, C. E. Phillips, Paolo Fraccaro, Denys Godwin, Bianca Zadrozny, Daniela Szwarcman, Carlos Gomes, Gabby Nyirjesy, Blair Edwards, Daiki Kimura, Naomi Simumba, Linsong Chu, S. Karthik Mukkavilli, Devyani Lambhate, Kamal Das, Ranjini Bangalore, Dario Oliveira, Michal Muszynski, Kumar Ankur, Muthukumaran Ramasubramanian, Iksha Gurung, Sam Khallaghi, Hanxi, Li, Michael Cecil, Maryam Ahmadi, Fatemeh Kordi, Hamed Alemohammad, Manil Maskey, Raghu Ganti, Kommy Weldemariam, Rahul Ramachandran
Significant progress in the development of highly adaptable and reusable Artificial Intelligence (AI) models is expected to have a significant impact on Earth science and remote sensing. Foundation models are pre-trained on large unlabeled datasets through self-supervision, and then fine-tuned for various downstream tasks with small labeled datasets. This paper introduces a first-of-a-kind framework for the efficient pre-training and fine-tuning of foundational models on extensive geospatial data. We have utilized this framework to create Prithvi, a transformer-based geospatial foundational model pre-trained on more than 1TB of multispectral satellite imagery from the Harmonized Landsat-Sentinel 2 (HLS) dataset. Our study demonstrates the efficacy of our framework in successfully fine-tuning Prithvi to a range of Earth observation tasks that have not been tackled by previous work on foundation models involving multi-temporal cloud gap imputation, flood mapping, wildfire scar segmentation, and multi-temporal crop segmentation. Our experiments show that the pre-trained model accelerates the fine-tuning process compared to leveraging randomly initialized weights. In addition, pre-trained Prithvi compares well against the state-of-the-art, e.g., outperforming a conditional GAN model in multi-temporal cloud imputation by up to 5pp (or 5.7%) in the structural similarity index. Finally, due to the limited availability of labeled data in the field of Earth observation, we gradually reduce the quantity of available labeled data for refining the model to evaluate data efficiency and demonstrate that data can be decreased significantly without affecting the model's accuracy. The pre-trained 100 million parameter model and corresponding fine-tuning workflows have been released publicly as open source contributions to the global Earth sciences community through Hugging Face.
Alexandre Lacoste, Nils Lehmann, Pau Rodriguez, Evan David Sherwin, Hannah Kerner, Björn Lütjens, Jeremy Andrew Irvin, David Dao, Hamed Alemohammad, Alexandre Drouin, Mehmet Gunturkun, Gabriel Huang, David Vazquez, Dava Newman, Yoshua Bengio, Stefano Ermon, Xiao Xiang Zhu
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.
Sam Khallaghi, Rahebe Abedi, Hanan Abou Ali, Hamed Alemohammad, Mary Dziedzorm Asipunu, Ismail Alatise, Nguyen Ha, Boka Luo, Cat Mai, Lei Song, Amos Wussah, Sitian Xiong, Yao-Ting Yao, Qi Zhang, Lyndon D. Estes
The accuracy of mapping agricultural fields across large areas is steadily improving with high-resolution satellite imagery and deep learning (DL) models, even in regions where fields are small and geometrically irregular. However, developing effective DL models often requires large, expensive label datasets, typically available only for specific years or locations. This limits the ability to create annual maps essential for agricultural monitoring, as domain shifts occur between years and regions due to changes in farming practices and environmental conditions. The challenge is to design a model flexible enough to account for these shifts without needing yearly labels. While domain adaptation techniques or semi-supervised training are common solutions, we explored enhancing the model's generalization power. Our results indicate that a holistic approach is essential, combining methods to improve generalization. Specifically, using an area-based loss function, such as Tversky-focal loss (TFL), significantly improved predictions across multiple years. The use of different augmentation techniques helped to encode different types of invariance, particularly photometric augmentations encoded invariance to brightness changes, though they increased false positives. The combination of photometric augmentation, TFL loss, and MC-dropout produced the best results, although dropout alone led to more false negatives in subsequent year predictions. Additionally, the choice of input normalization had a significant impact, with the best results obtained when statistics were calculated either locally or across the entire dataset over all bands (lab and gab). We developed a workflow that enabled a U-Net model to generate effective multi-year crop maps over large areas. Our code, available at: https://github.com/agroimpacts/cnn-generalization-enhancement, will be regularly updated with improvements.
Mirela G. Tulbure, Julio Caineta, Mark Broich, Mollie D. Gaines, Philippe Rufin, Leon-Friedrich Thomas, Hamed Alemohammad, Jan Hemmerling, Patrick Hostert
Floods are among the most damaging weather-related hazards, and in 2024, the warmest year on record, extreme flood events affected communities across five continents. Earth observation (EO) satellites provide critical, frequent coverage for mapping inundation, yet operational accuracy depends heavily on labeled datasets and model generalization. Recent Geospatial Foundation Models (GFMs), such as ESA-IBM's TerraMind, offer improved generalizability through large-scale self-supervised pretraining, but their performance on diverse global flood events remains poorly understood. We fine-tune TerraMind for flood extent mapping using FloodsNet, a harmonized multimodal dataset containing co-located Sentinel-1 (Synthetic Aperture Radar, SAR data) and Sentinel-2 (optical) imagery for 85 flood events worldwide. We tested four configurations (base vs. large models; frozen vs. unfrozen backbones) and compared against the TerraMind Sen1Floods11 example and a U-Net trained on both FloodsNet and Sen1Floods11. The base-unfrozen configuration provided the best balance of accuracy, precision, and recall at substantially lower computational cost than the large model. The large unfrozen model achieved the highest recall. Models trained on FloodsNet outperformed the Sen1Floods11-trained example in recall with similar overall accuracy. U-Net achieved higher recall than all GFM configurations, though with slightly lower accuracy and precision. Our results demonstrate that integrating multimodal optical and SAR data and fine-tuning a GFM can enhance near-real-time flood mapping. This study provides one of the first global-scale evaluations of a GFM for flood segmentation, highlighting both its potential and current limitations for climate adaptation and disaster resilience.