The scarcity of labeled satellite imagery remains a fundamental bottleneck for deep-learning (DL)-based wildfire monitoring systems. This paper investigates whether a diffusion-based foundation model for Earth Observation (EO), EarthSynth, can synthesize realistic post-wildfire Sentinel-2 RGB imagery conditioned on existing burn masks, without task-specific retraining.
Using burn masks derived from the CalFireSeg-50 dataset (Martin et al., 2025), we design and evaluate six controlled experimental configurations that systematically vary: (i) pipeline architecture (mask-only full generation vs. inpainting with pre-fire context), (ii) prompt engineering strategy (three hand-crafted prompts and a VLM-generated prompt via Qwen2-VL), and (iii) a region-wise color-matching post-processing step.
Quantitative assessment on 10 stratified test samples uses four complementary metrics: Burn IoU, burn-region color distance (ΔC_burn), Darkness Contrast, and Spectral Plausibility. Results show that inpainting-based pipelines consistently outperform full-tile generation across all metrics, with the structured inpainting prompt achieving the best spatial alignment (Burn IoU = 0.456) and burn saliency (Darkness Contrast = 20.44), while color matching produces the lowest color distance (ΔC_burn = 63.22) at the cost of reduced burn saliency.
VLM-assisted inpainting is competitive with hand-crafted prompts. These findings provide a foundation for incorporating generative data augmentation into wildfire detection pipelines.
Code and experiments are available at: https://www.kaggle.com/code/valeriamartinh/genai-all-runned
Wildfires have increased in frequency and severity over the past two decades, especially in the Western United States. Beyond physical infrastructure damage caused by these wildfire events, researchers have increasingly identified harmful impacts of particulate matter generated by wildfire smoke on respiratory, cardiovascular, and cognitive health. This inference is difficult due to the spatial and temporal uncertainty regarding how much particulate matter is specifically attributable to wildfire smoke. One factor contributing to this challenge is the reliance on manually drawn smoke plume ann…
FLAME 3 Dataset: Unleashing the Power of Radiometric Thermal UAV Imagery for Wildfire Management
Holcombe Department of Electrical and Computer Engineering, Clemson University（克莱姆森大学霍尔科姆电气与计算机工程系）；Pacific Southwest Research Station, U.S. Forest Services（美国林业局太平洋西南研究站）；School of Environmental and Forest Sciences, University of Washington（华盛顿大学环境与森林科学学院）；US Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory（美国林业局落基山研究站火灾科学实验室）；Department of Mechanical Engineering, Clemson University（克莱姆森大学机械工程系）；Department of Forest Management, University of Montana（蒙大拿大学森林管理系）
Since frequent severe droughts are lengthening the dry season in the Amazon Rainforest, it is important to detect wildfires promptly and forecast possible spread for effective suppression response. Current wildfire detection models are not versatile enough for the low-technology conditions of South American hot spots. This deep learning study first trains a Fully Convolutional Neural Network on Landsat 8 images of Ecuador and the Galapagos, using Green and Short-wave Infrared bands to predict pixel-level binary fire masks. This model achieves a 0.962 validation F2 score and a 0.932 F2 score on…
Spatio-Temporal Wildfire Spread Prediction in Canada using a Video Swin-Hybrid-U-Net and Satellite Imagery
Faculty of Science, University of Alberta（阿尔伯塔大学理学院）；Interdisciplinary Lab for Mathematical Ecology & Epidemiology (ILMEE), University of Alberta（阿尔伯塔大学数学生态与流行病学跨学科实验室）；Department of Mathematical and Statistical Sciences, University of Alberta（阿尔伯塔大学数学与统计科学系）
Harnessing Self-Supervised Deep Learning and Geostationary Remote Sensing for Advancing Wildfire and Associated Air Quality Monitoring: Improved Smoke and Fire Front Masking using GOES and TEMPO Radiance Data
Spatial Informatics Group, LLC.（空间信息集团有限责任公司）；ADNET systems, INC.（ADNET系统公司）；NASA Langley Research Center（美国国家航空航天局兰利研究中心）；University at Albany（奥尔巴尼大学）
Orion Lab, National Observatory of Athens & National Technical University of Athens（Orion实验室、雅典国家天文观测站及雅典技术大学）；Department of Informatics and Telematics, Harokopio University of Athens（信息与电信系、哈罗基奥大学）
Università di Corsica（科西嘉大学）；Tecnológico de Monterrey（蒙特雷理工学院）
Early wildfire detection is of paramount importance to avoid as much damage as possible to the environment, properties, and lives. Deep Learning (DL) models that can leverage both visible and infrared information have the potential to display state-of-the-art performance, with lower false-positive rates than existing techniques. However, most DL-based image fusion methods have not been evaluated in the domain of fire imagery. Additionally, to the best of our knowledge, no publicly available dataset contains visible-infrared fused fire images. There is a growing interest in DL-based image fusio…
FireSentry: A Multi-Modal Spatio-temporal Benchmark Dataset for Fine-Grained Wildfire Spread Forecasting
Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Department of Electronic Engineering, Tsinghua University（清华大学电子工程系）；College of Soil and Water Conservation, Southwest Forestry University（西南林业大学水土保持学院）；College of Civil Engineering, Southwest Forestry University（西南林业大学土木工程学院）
Copelabs, Lusófona University（Copelabs 和 奥埃拉法纳大学）；Center of Technology and Systems (UNINOVA-CTS) and Associated Lab of Intelligent Systems (LASI)（技术与系统中心（UNINOVA-CTS）及智能系统联合实验室）；NOVA School of Science and Technology, NOVA University Lisbon（科学与技术学院（NOVA大学里斯本））
Annual Earth-observation embeddings encode wildfire disturbance and support simplified burned area mapping
Conservation Research Institute, University of Cambridge（剑桥大学保护研究所）；Department of Plant Sciences, University of Cambridge（剑桥大学植物科学系）；dClimate Labs（dClimate实验室）；Department of Computer Science and Technology, University of Cambridge（剑桥大学计算机科学与技术系）；Department of Earth System Science, Stanford University（斯坦福大学地球系统科学系）
An FPGA smart camera implementation of segmentation models for drone wildfire imagery
Tecnologico de Monterrey（蒙特雷科技大学）；Hasso-Plattner Institute, University of Postdam（波茨坦大学哈索·普拉特纳研究所）；Universidad Autonoma de Guadalajara（瓜达拉哈拉自治大学）；Universidad de las Americas Puebla（普埃布拉美洲大学）；Università di Corsica（科西嘉大学）
A Large Scale Open-Source Image and Video Dataset for Robust Wildfire Detection and Classification
Department of Electrical and Computer Engineering, University of Illinois Chicago（伊利诺伊大学芝加哥分校电子与计算机工程系）；Informatics Institute, Istanbul Technical University（伊斯坦布尔技术大学信息学院）；USDA Forest Service Pacific Wildland Fire Sciences Laboratory（美国农业部森林服务局太平洋野火科学实验室）；Department of Computer Engineering, Bilkent University（比尔肯特大学计算机工程系）
Institute for Applied Computational Science, Harvard University（哈佛大学应用计算科学研究所）；Microsoft AI for Good Research Lab（微软人工智能向善研究实验室）
In this paper we propose a mask-conditional synthetic image generation model for creating synthetic satellite imagery datasets. Given a dataset of real high-resolution images and accompanying land cover masks, we show that it is possible to train an upstream conditional synthetic imagery generator, use that generator to create synthetic imagery with the land cover masks, then train a downstream model on the synthetic imagery and land cover masks that achieves similar test performance to a model that was trained with the real imagery. Further, we find that incorporating a mixture of real and sy…
Skolkovo institute of Science and technology（斯科尔科沃科学技术研究院）；Moscow（莫斯科）
When major disaster occurs the questions are raised how to estimate the damage in time to support the decision making process and relief efforts by local authorities or humanitarian teams. In this paper we consider the use of Machine Learning and Computer Vision on remote sensing imagery to improve time efficiency of assessment of damaged buildings in disaster affected area. We propose a general workflow that can be useful in various disaster management applications, and demonstrate the use of the proposed workflow for the assessment of the damage caused by the wildfires in California in 2017.
Google Research（谷歌研究院）；Stanford University（斯坦福大学）；United Nations World Food Programme（联合国世界粮食计划署）
To respond to disasters such as earthquakes, wildfires, and armed conflicts, humanitarian organizations require accurate and timely data in the form of damage assessments, which indicate what buildings and population centers have been most affected. Recent research combines machine learning with remote sensing to automatically extract such information from satellite imagery, reducing manual labor and turn-around time. A major impediment to using machine learning methods in real disaster response scenarios is the difficulty of obtaining a sufficient amount of labeled data to train a model for a…
FUELVISION: A Multimodal Data Fusion and Multimodel Ensemble Algorithm for Wildfire Fuels Mapping
University of California Los Angeles（加利福尼亚大学洛杉矶分校）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Desert Research Institute（沙漠研究所）；Amazon（亚马逊公司）；United States Forest Service（美国林务局）
本文提出 FUELVISION 算法，融合多源遥感与地形数据，结合生成式 AI 合成数据增强训练集，采用多模型集成方法实现近实时野火可燃物制图，准确率近 80%。
University of Nebraska-Lincoln（内布拉斯加大学林肯分校）；University of California, Berkeley（加州大学伯克利分校）；Boston University（波士顿大学）；Lisbon University（里斯本大学）；Northwest Nazarene University（西北拿撒勒大学）；University of Oxford（牛津大学）
The accurate characterization of the severity of the wildfire event strongly contributes to the characterization of the fuel conditions in fire-prone areas, and provides valuable information for disaster response. The aim of this study is to develop an autonomous system built on top of high-resolution multispectral satellite imagery, with an advanced deep learning method for detecting burned area change. This work proposes an initial exploration of using an unsupervised model for feature extraction in wildfire scenarios. It is based on the contrastive learning technique SimCLR, which is traine…
University of California, San Diego（加州大学圣迭戈分校）；San Diego Supercomputer Center（圣迭戈超级计算机中心）；Scripps Institution of Oceanography（斯克里普斯海洋研究所）
The size and frequency of wildland fires in the western United States have dramatically increased in recent years. On high-fire-risk days, a small fire ignition can rapidly grow and become out of control. Early detection of fire ignitions from initial smoke can assist the response to such fires before they become difficult to manage. Past deep learning approaches for wildfire smoke detection have suffered from small or unreliable datasets that make it difficult to extrapolate performance to real-world scenarios. In this work, we present the Fire Ignition Library (FIgLib), a publicly available …
Improving Satellite Imagery Masking using Multi-task and Transfer Learning
College of Information and Computer Sciences, University of Massachusetts（马萨诸塞大学信息与计算机科学学院）；University of Pittsburgh（匹兹堡大学）；University of North Carolina（北卡罗来纳大学）
Unsupervised pre-training methods for large vision models have shown to enhance performance on downstream supervised tasks. Developing similar techniques for satellite imagery presents significant opportunities as unlabelled data is plentiful and the inherent temporal and multi-spectral structure provides avenues to further improve existing pre-training strategies. In this paper, we present SatMAE, a pre-training framework for temporal or multi-spectral satellite imagery based on Masked Autoencoder (MAE). To leverage temporal information, we include a temporal embedding along with independentl…
The parsing of windows in building facades is a long-desired but challenging task in computer vision. It is crucial to urban analysis, semantic reconstruction, lifecycle analysis, digital twins, and scene parsing amongst other building-related tasks that require high-quality semantic data. This article investigates the usage of the mask R-CNN framework to be used for window detection of facade imagery input. We utilize transfer learning to train our proposed method on COCO weights with our own collected dataset of street view images of facades to produce instance segmentations of our new windo…
Booz Allen Hamilton（博思艾伦汉密尔顿）；National Geospatial-Intelligence Agency（美国国家地理空间情报局）
Labeling data to use for training object detectors is expensive and time consuming. Publicly available overhead datasets for object detection are labeled with image-aligned bounding boxes, object-aligned bounding boxes, or object masks, but it is not clear whether such detailed labeling is necessary. To test the idea, we developed novel single- and two-stage network architectures that use centerpoints for labeling. In this paper we show that these architectures achieve nearly equivalent performance to approaches using more detailed labeling on three overhead object detection datasets.
This paper studies the problem of polygonal mapping of buildings by tackling the issue of mask reversibility that leads to a notable performance gap between the predicted masks and polygons from the learning-based methods. We addressed such an issue by exploiting the hierarchical supervision (of bottom-level vertices, mid-level line segments and the high-level regional masks) and proposed a novel interaction mechanism of feature embedding sourced from different levels of supervision signals to obtain reversible building masks for polygonal mapping of buildings. As a result, we show that the le…
NASA Ames Research Center（美国国家航空航天局艾姆斯研究中心）；Bay Area Environmental Research Institute（湾区环境研究所）
Applications of satellite data in areas such as weather tracking and modeling, ecosystem monitoring, wildfire detection, and land-cover change are heavily dependent on the trade-offs to spatial, spectral and temporal resolutions of observations. In weather tracking, high-frequency temporal observations are critical and used to improve forecasts, study severe events, and extract atmospheric motion, among others. However, while the current generation of geostationary satellites have hemispheric coverage at 10-15 minute intervals, higher temporal frequency observations are ideal for studying meso…
Enhancing deep learning performance on burned area delineation from SPOT-6/7 imagery for emergency management
IRISA, Université Bretagne Sud, UMR 6074, Vannes, France（IRISA，布列塔尼大学，UMR 6074，瓦纳兹，法国）；University of Salzburg, Department of Geoinformatics, Salzburg, Austria（萨尔茨堡大学，地理信息学院，萨尔茨堡，奥地利）；ICube-SERTIT, Université de Strasbourg, Illkirch Graffenstaden, France（ICube-SERTIT，斯特拉斯堡大学，伊尔克里希格拉芬施塔登，法国）
COMSATS University Islamabad（伊斯兰堡COMSATS大学）；University of Tasmania（塔斯马尼亚大学）；National University of Ireland, Galway（爱尔兰国立高威大学）
Food resources face severe damages under extraordinary situations of catastrophes such as earthquakes, cyclones, and tsunamis. Under such scenarios, speedy assessment of food resources from agricultural land is critical as it supports aid activity in the disaster hit areas. In this article, a deep learning approach is presented for the detection and segmentation of coconut tress in aerial imagery provided through the AI competition organized by the World Bank in collaboration with OpenAerialMap and WeRobotics. Maked Region-based Convolutional Neural Network approach was used identification and…
Universitat de València（瓦伦西亚大学）；Descartes Lab（笛卡尔实验室）
Anomalous change detection (ACD) is an important problem in remote sensing image processing. Detecting not only pervasive but also anomalous or extreme changes has many applications for which methodologies are available. This paper introduces a nonlinear extension of a full family of anomalous change detectors. In particular, we focus on algorithms that utilize Gaussian and elliptically contoured (EC) distribution and extend them to their nonlinear counterparts based on the theory of reproducing kernels' Hilbert space. We illustrate the performance of the kernel methods introduced in both perv…
School of Resource and Environmental Sciences, Wuhan University（武汉大学资源与环境科学学院）；Collaborative Innovation Center for Geospatial Technology（地球空间信息技术协同创新中心）；Key Laboratory of Geographic Information System, Ministry of Education, Wuhan University（武汉大学地理信息系统教育部重点实验室）；State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University（武汉大学测绘遥感信息工程国家重点实验室）；Department of Electronics, University of Pavia（帕维亚大学电子系）
The wide field of view (WFV) imaging system onboard the Chinese GaoFen-1 (GF-1) optical satellite has a 16-m resolution and four-day revisit cycle for large-scale Earth observation. The advantages of the high temporal-spatial resolution and the wide field of view make the GF-1 WFV imagery very popular. However, cloud cover is an inevitable problem in GF-1 WFV imagery, which influences its precise application. Accurate cloud and cloud shadow detection in GF-1 WFV imagery is quite difficult due to the fact that there are only three visible bands and one near-infrared band. In this paper, an auto…
Detection of Endangered Deer Species Using UAV Imagery: A Comparative Study Between Efficient Deep Learning Approaches
Artificial Intelligence and Robotics Laboratory (LINAR)（人工智能与机器人实验室）；Universidad de San Andrés（圣安德烈大学）；Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)（国家科学与技术研究中心）；Centre for Automation and Robotics (CSIC-UPM)（自动化与机器人中心）；Museo Argentino Bernardino Rivadavia（阿根廷Bernardino Rivadavia博物馆）
