On the Generalizability of Foundation Models for Crop Type Mapping
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Yi-Chia Chang, Adam J. Stewart, Favyen Bastani, Piper Wolters, Shreya Kannan, George R. Huber, Jingtong Wang, Arindam Banerjee
备注：
Accepted to IEEE IGARSS 2025. The final version is available in the Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2025
Foundation models pre-trained using self-supervised learning have shown powerful transfer learning capabilities on various downstream tasks, including language understanding, text generation, and image recognition. The Earth observation (EO) field has produced several foundation models pre-trained directly on multispectral satellite imagery for applications like precision agriculture, wildfire and drought monitoring, and natural disaster response. However, few studies have investigated the ability of these models to generalize to new geographic locations, and potential concerns of geospatial bias -- models trained on data-rich developed nations not transferring well to data-scarce developing nations -- remain. We evaluate three popular EO foundation models, SSL4EO-S12, SatlasPretrain, and ImageNet, on five crop classification datasets across five continents. Results show that pre-trained weights designed explicitly for Sentinel-2, such as SSL4EO-S12, outperform general pre-trained weights like ImageNet. While only 100 labeled images are sufficient for achieving high overall accuracy, 900 images are required to mitigate class imbalance and improve average accuracy.
Can SAM recognize crops? Quantifying the zero-shot performance of a semantic segmentation foundation model on generating crop-type maps using satellite imagery for precision agriculture
SPROUT: A Scalable Diffusion Foundation Model for Multi-Crop Plant Phenotyping
Graduate School of Agricultural and Life Sciences, The University of Tokyo（东京大学大学院农学生命科学研究科）；National Key Laboratory of Multispectral Information Intelligent Processing Technology, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学多谱信息智能处理技术全国重点实验室、人工智能与自动化学院）；Institute of Agricultural Machinery, NARO（日本国立研究开发法人农业·食品产业技术综合研究机构农业机械研究所）；Institute of Life and Environmental Sciences, University of Tsukuba（筑波大学生命环境科学研究所）
Can Time-Series Foundation Models Perform Building Energy Management Tasks?
Carnegie Mellon University（卡内基梅隆大学）；University of California, Los Angeles（加州大学洛杉矶分校）；Hong Kong University of Science and Technology（香港科学与技术大学）；Amazon（亚马逊公司）
Benchmarking foundation models for hyperspectral image classification: Application to cereal crop type mapping
Analytics Lab (A-Lab), UM6P, Rabat 11103, Morocco（安利实验室（A-Lab））；Center for Remote Sensing Applications (CRSA), UM6P, Ben Guerir 43150, Morocco（远程传感应用中心（CRSA））；Friedrich Schiller University Jena, Department of Geography, Jena 07743, Germany（耶拿弗里德里希·施勒格尔大学）；Dept. of Cartography（制图部门）
DepthCropSeg++: Scaling a Crop Segmentation Foundation Model With Depth-Labeled Data
MetaPheno Laboratory（MetaPheno实验室）；PhenoTrait Technology Co., Ltd.（PhenoTrait技术有限公司）；Wuhan Digital Engineering Institute（武汉数字工程研究院）；National Key Laboratory of Multispectral Information Intelligent Processing Technology（国家多光谱信息智能处理技术重点实验室）；School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学人工智能与自动化学院）；Hubei Key Laboratory of Intelligent Robot（湖北省智能机器人重点实验室）；School of Computer Science and Engineering, School of Artificial Intelligence, Wuhan Institute of Technology（武汉理工大学计算机科学与工程学院、人工智能学院）
FoMo4Wheat: Toward reliable crop vision foundation models with globally curated data
Nanjing Agricultural University（南京农业大学）；Huazhong University of Science and Technology（华中科技大学）；Beijing University of Posts and Telecommunications（北京邮电大学）；Northwest A&F University（西北农林科技大学）；The University of Queensland（昆士兰大学）；Commonwealth Scientific and Industrial Research Organization（澳大利亚联邦科学与工业研究组织）；The University of Tokyo（东京大学）；INRAE（法国国家农业食品与环境研究院）；Arvalis（阿瓦利斯植物研究院）；ETH Zurich（苏黎世联邦理工学院）；International Center for Agricultural Research in the Dry Areas (ICARDA)（国际干旱地区农业研究中心）；University of Nebraska-Lincoln（内布拉斯加大学林肯分校）；Chinese Academy of Agricultural Sciences（中国农业科学院）；Hiphen SAS（Hiphen公司）
AgriFM: A Multi-source Temporal Remote Sensing Foundation Model for Agriculture Mapping
Jockey Club STEM Lab of Quantitative Remote Sensing, Department of Geography, The University of Hong Kong, Hong Kong, China（香港大学地理系金钟STEM实验室）；Department of Aerospace Intelligent Science and Technology, School of Astronautics, Beihang University, Beijing, China（北航航天智能科学与技术学院）；School of Remote Sensing and Information Engineering, Wuhan University, China（武汉大学遥感与信息工程学院）
The University of Tokyo（东京大学）；National Institute of Informatics（国立信息学研究所）；Microsoft Research（微软研究院）
Image cropping has progressed tremendously under the data-driven paradigm. However, current approaches do not account for the intentions of the user, which is an issue especially when the composition of the input image is complex. Moreover, labeling of cropping data is costly and hence the amount of data is limited, leading to poor generalization performance of current algorithms in the wild. In this work, we take advantage of vision-language models as a foundation for creating robust and user-intentional cropping algorithms. By adapting a transformer decoder with a pre-trained CLIP-based dete…
Segment Anything Model Can Not Segment Anything: Assessing AI Foundation Model's Generalizability in Permafrost Mapping
Arizona State University（亚利桑那州立大学）；Woodwell Climate Research Center（伍德韦尔气候研究中心）；University of Connecticut（康涅狄格大学）；U.S. Geological Survey（美国地质调查局）；University of California, Santa Barbara（加利福尼亚大学圣巴巴拉分校）；University of Illinois at Urbana-Champaign（伊利诺伊大学香槟分校）
本文评估了以 Meta SAM 为代表的视觉基础模型在永久冻土地貌分割中的表现，通过多种提示策略和微调测试其零样本能力与领域适应性，并指出其在挑战性地理空间任务中仍需改进。
Bias and Generalizability of Foundation Models across Datasets in Breast Mammography
University Hospital Bonn（波恩大学医院）；Technical University of Munich（慕尼黑技术大学）；Lebanese Hospital Geitaoui（黎巴嫩Geitaoui医院）；Helmholtz AI, Helmholtz Munich（海德堡人工智能，海德堡慕尼黑）
Training real-world neural network models to achieve high performance and generalizability typically requires a substantial amount of labeled data, spanning a broad range of variation. This data-labeling process can be both labor and cost intensive. To achieve desirable predictive performance, a trained model is typically applied into a domain where the data distribution is similar to the training dataset. However, for many agricultural machine learning problems, training datasets are collected at a specific location, during a specific period in time of the growing season. Since agricultural s…
Crop Pest Classification Using Deep Learning Techniques: A Review
FAST School of Computing, National University of Computer and Emerging Sciences（FAST计算学院，国立计算机与新兴科学大学）；Department of Artificial Intelligence, Ajou University（人工智能系，阿乔大学）
California Crop Yield Benchmark: Combining Satellite Image, Climate, Evapotranspiration, and Soil Data Layers for County-Level Yield Forecasting of Over 70 Crops
University of California, Davis（加州大学戴维斯分校）；Texas A&M University-Corpus Christi（德克萨斯A&M大学-科斯蒂西）；AI Institute for Food Systems（人工智能食品系统研究所）
Landslide Hazard Mapping with Geospatial Foundation Models: Geographical Generalizability, Data Scarcity, and Band Adaptability
School of Geographical Sciences and Urban Planning（地理科学与城市规划学院）；Arizona State University（亚利桑那州立大学）；School of Computing and Augmented Intelligence（计算与增强智能学院）；NASA Marshall Space Flight Center（国家航空航天局马歇尔航天飞行中心）
ViTime: Foundation Model for Time Series Forecasting Powered by Vision Intelligence
Technion – Israel Institute of Technology（以色列理工学院）；Manchester Metropolitan University（曼彻斯特城市大学）；Northwest Polytechnical University（西北工业大学）；City University of Hong Kong（香港城市大学）
Adapting Foundation Model for Dental Caries Detection with Dual-View Co-Training
School of Biomedical Engineering \& State Key Laboratory of Advanced Medical Materials；Devices, ShanghaiTech University, Shanghai, China Lingang Laboratory, Shanghai, China Shanghai United Imaging Intelligence Co. Ltd., Shanghai, China Shanghai Clinical Research；Trial Center, Shanghai, China Shanghai Linkedcare Information Technology Co., Ltd., Shanghai, China
Rethinking Foundation Models for Medical Image Classification through a Benchmark Study on MedMNIST
University of Oxford（牛津大学）；Nuffield Department of Population Health, University of Oxford（牛津大学纳菲尔德人口健康系）；Big Data Institute, University of Oxford（牛津大学大数据研究所）
Tissue Concepts: supervised foundation models in computational pathology
Fraunhofer Institute for Digital Medicine MEVIS（弗劳恩霍夫数字医学研究所MEVIS）；Institute for Pathology, Hannover Medical School（汉诺威医学院病理学研究所）；University of Regensburg（雷根斯堡大学）；RWTH Aachen University（亚琛工业大学）
TotalRegistrator: Towards a Lightweight Foundation Model for CT Image Registration
Department of Imaging, Radboudumc, Nijmegen, the Netherlands（影像部，Radboudumc，尼姆egen，荷兰）；Institute for Medical Informatics, University of Lübeck, Lübeck, Germany（医学信息研究所，吕贝克大学，吕贝克，德国）；Department of Nuclear Medicine, Hospital 108, Hanoi, Vietnam（核医学部，108医院，河内，越南）；Department of Diagnostic Imaging and Interventional Radiology, Thai Nguyen National Hospital, Thai Nguyen, Vietnam（诊断影像和介入放射科，Thai Nguyen国家医院，Thai Nguyen，越南）；Interventional Radiology Center, Tam Anh Hospital, Hanoi, Vietnam（介入放射科中心，Tam Anh医院，河内，越南）；FET, Vietnam National University, University of Engineering and Technology, Hanoi, Vietnam（FET，越南国家大学，工程与技术大学，河内，越南）
