Very-High Resolution (VHR) remote sensing imagery is increasingly accessible, but often lacks annotations for effective machine learning applications. Recent foundation models like GroundingDINO [1] and Segment Anything (SAM) [2] provide opportunities to automatically generate annotations. This study introduces FMARS (Foundation Model Annotations in Remote Sensing), a methodology leveraging VHR imagery and foundation models for fast and robust annotation. We focus on disaster management and provide a large-scale dataset with labels obtained from pre-event imagery over 19 disaster events, derived from the Maxar Open Data initiative. We train segmentation models on the generated labels, using Unsupervised Domain Adaptation (UDA) techniques to increase transferability to real-world scenarios. Our results demonstrate the effectiveness of leveraging foundation models to automatically annotate remote sensing data at scale, enabling robust downstream models for critical applications. Code and dataset are available at https://github.com/links-ads/igarss-fmars.

Rapid detection and well-timed intervention are essential to mitigate the impacts of wildfires. Leveraging remote sensed data from satellite networks and advanced AI models to automatically detect hotspots (i.e., thermal anomalies caused by active fires) is an effective way to build wildfire monitoring systems. In this work, we propose a novel dataset containing time series of remotely sensed data related to European fire events and a Self-Supervised Learning (SSL)-based model able to analyse multi-temporal data and identify hotspots in potentially near real time. We train and evaluate the performance of our model using our dataset and Thraws, a dataset of thermal anomalies including several fire events, obtaining an F1 score of 63.58.

Landslides are one of the most critical and destructive geohazards. Widespread development of human activities and settlements combined with the effects of climate change on weather are resulting in a high increase in the frequency and destructive power of landslides, making them a major threat to human life and the economy. In this paper, we explore methodologies to map newly-occurred landslides using Sentinel-2 imagery automatically. All approaches presented are framed as a bi-temporal change detection problem, requiring only a pair of Sentinel-2 images, taken respectively before and after a landslide-triggering event. Furthermore, we introduce a novel deep learning architecture for fusing Sentinel-2 bi-temporal image pairs with Digital Elevation Model (DEM) data, showcasing its promising performances w.r.t. other change detection models in the literature. As a parallel task, we address limitations in existing datasets by creating a novel geodatabase, which includes manually validated open-access landslide inventories over heterogeneous ecoregions of the world. We release both code and dataset with an open-source license.

Land cover (LC) segmentation plays a critical role in various applications, including environmental analysis and natural disaster management. However, generating accurate LC maps is a complex and time-consuming task that requires the expertise of multiple annotators and regular updates to account for environmental changes. In this work, we introduce SPADA, a framework for fuel map delineation that addresses the challenges associated with LC segmentation using sparse annotations and domain adaptation techniques for semantic segmentation. Performance evaluations using reliable ground truths, such as LUCAS and Urban Atlas, demonstrate the technique’s effectiveness. SPADA outperforms state-of-the-art semantic segmentation approaches as well as third-party products, achieving a mean Intersection over Union (IoU) score of 42.86 and an F1 score of 67.93 on Urban Atlas and LUCAS, respectively.

The increasing frequency of catastrophic natural events, such as wildfires, calls for the development of rapid and automated wildfire detection systems. In this paper, we propose a wildfire identification solution to improve the accuracy of automated satellite-based hotspot detection systems by leveraging multiple information sources. We cross-reference the thermal anomalies detected by the Moderate-resolution Imaging Spectroradiometer (MODIS) and the Visible Infrared Imaging Radiometer Suite (VIIRS) hotspot services with the European Forest Fire Information System (EFFIS) database to construct a large-scale hotspot dataset for wildfire-related studies in Europe. Then, we propose a novel multimodal supervised machine learning approach to disambiguate hotspot detections, distinguishing between wildfires and other events. Our methodology includes the use of multimodal data sources, such as the ERSI annual Land Use Land Cover (LULC) and the Copernicus Sentinel-3 data. Experimental results demonstrate the effectiveness of our approach in the task of wildfire identification.

In the last years, we are witnessing an increasing availability of geolocated data, ranging from satellite images to user generated content (e.g., tweets). This big amount of data is exploited by several cloud-based applications to deliver effective and customized services to end users. In order to provide a good user experience, a low-latency response time is needed, both when data are retrieved and provided. To achieve this goal, current geospatial applications need to exploit efficient and scalable geospatial databases, the choice of which has a high impact on the overall performance of the deployed applications. In this paper, we compare, from a qualitative point of view, four state-of-the-art SQL and NoSQL databases with geospatial features, and then we analyze the performances of two of them, selecting the ones based on the Database-as-a-service (DBaaS) model: Azure SQL Database and Azure DocumentDB (i.e., an SQL database versus a NoSQL one). The empirical evaluation shows pros and cons of both solutions and it is performed on a real use case related to an emergency management application.

The threat of a forest fire disaster increases around the globe as the human footprint continues to encroach on natural areas and climate change effects increase the potential of extreme weather. It is essential that the tools to educate, prepare, monitor, react, and fight natural fire disasters are available to emergency managers and responders and reduce the overall disaster effects. In the context of the I-REACT project, such a big crisis data system is being developed and is based on the integration of information from different sources, automated data processing chains and decision support systems. This paper presents the wildfire monitoring for emergency management system for those involved and affected by wildfire disasters developed for European forest fire disasters.

Nowadays, the number and the magnitude of natural hazards are increasing. During emergency situations different forms of cooperation take place, including the crowd-sourcing, which is envisioned by many Disaster Risk Management approaches to enable citizens to support the emergency management process. However, crowdsourcing is a challenging paradigm as it requires a sustained engagement in order to be effective. In this paper we propose a gamification strategy for crowdsourced Disaster Risk Management services aimed to increase awareness, engagement, and change people behaviors.

In this paper, we propose a machine learning approach to automatically classify non-informative and informative contents shared on Twitter during disasters caused by natural hazards. In particular, we leverage on previously sampled and labeled datasets of messages posted on Twitter during or in the aftermath of natural disasters. Starting from results obtained in previous studies, we propose a language-agnostic model. We define a base feature set considering only Twitter-specific metadata of each tweet, using classification results from this set as a reference. We introduce an additional feature, called the Source Feature, which is computed considering the device or platform used to post a tweet, and we evaluate its contribution in improving the classifier accuracy.

Floods are major natural disasters which cause deaths and material damages every year. Monitoring these events is crucial in order to reduce both the affected people and the economic losses. In this work we train and test three different Deep Learning segmentation algorithms to estimate the water area from river images, and compare their performances. We discuss the implementation of a novel data chain aimed to monitor river water levels by automatically process data collected from surveillance cameras, and to give alerts in case of high increases of the water level or flooding. We also create and openly publish the first image dataset for river water segmentation.

Summary Despite the growing development of space‐based systems aimed at monitoring and studying natural hazards, they continue to harm humankind worldwide, causing enormous human and economic losses. In a view of improving the effectiveness and the timeliness of existing emergency systems, we propose a service‐oriented cloud‐based software architecture for mobile sensing applications. Exploiting existing Global Navigation Satellite Systems and public Cloud Computing services, we implement a set of mobile geolocated services enabling mobile devices to send real‐time in‐field observations. Such observations could be used by authorities and first responders for both Early Warning and Emergency Response services complementing in real time the situational assessment provided by existing means, eg, remote sensed information and in‐situ sensors. We propose a Service Oriented Architecture that can be used as a reference for implementing the back end of mobile applications requiring to send crowdsourced geolocated reports. We fully implement a real application, and we evaluate its performances with Microsoft Azure varying the user load and the main deployment parameters. Our results can be taken as reference to assess the capability of future applications with similar requirements.

In this paper we propose, implement and evaluate a bandwidth aggregation service for residential users that enhances the throughput of their Internet broadband connection through the aggregation of available capacity at neighboring broadband links. Network resources are aggregated by the residential access gateway using the 802.11 radio interface to simultaneously serve home users and to share the broadband connectivity with neighboring access gateways. Differently from previous works, our aggregation scheme is transparent both for local users, who are not required to modify their applications or device drivers, and for neighboring users, who do not experience any meaningful performance degradation. The proposed approach aims at a commercial deployment, leveraging on existing access gateways and ADSL-based access networks.

Despite the growing development of space-based systems aimed at monitoring and studying natural hazards, floods continue to harm humankind worldwide, causing enormous human and economic losses. In view of improving the existing Copernicus Emergency Management Service (EMS) we propose FLOODIS: a novel Copernicus downstream service that exploits existing space assets together with crowdsourcing and state-of-the-art cloud computing technologies in order to provide a fast, flexible and scalable flood emergency service. After collecting the requirements from end-users, we define the FLOODIS architecture and implement a FLOODIS prototype. We also study the challenges related to consider real-time data coming from modern social networks like Twitter. Finally, we draft a business case for FLOODIS by analyzing the potential market along with the estimated social benefits brought by the proposed service.

The proliferation of overlapping, always-on IEEE 802.11 access points (APs) in urban areas, can cause inefficient bandwidth usage and energy waste. Cooperation among APs could address these problems by allowing underused devices to hand over their wireless stations to nearby APs and temporarily switch off, while avoiding to overload a BSS and thus offloading congested APs. The federated house model provides an appealing backdrop to implement cooperation among APs. In this paper, we outline a distributed framework that assumes the presence of a multipurpose gateway with AP capabilities in every household. Our framework allows cooperation through the monitoring of local wireless resources and the triggering of offloading requests toward other federated gateways. Our simulation results show that, in realistic residential settings, the proposed framework yields an energy saving between 45 and 86 percent under typical usage patterns, while avoiding congestion and meeting user expectations in terms of throughput. Furthermore, we show the feasibility and the benefits of our framework with a real test-bed deployed on commodity hardware.

Cooperative caching schemes allow to improve the performance of multi-hop networks based on device-to-device (D2D) communications. Indeed, each node does not only share its transmission capabilities to physically extend the network, but it also shares its storage to cache copies of contents for the sake of other nodes. It results in an increased network performance for users, since caching decreases both network load and latency to reach a content. The design of effective caching policies in a network of caches is very challenging and all the known solutions must be adapted both to the topology and to the request traffic pattern. In this paper, we consider a linear topology, representing a sequence of adjacent nodes, investigating the performances of both local and distributed cooperative caching policies. We specifically investigate where to apply the caching policy. Interestingly, we show that a simple local caching policy, that caches only the contents requested by the node itself, is not worse (or even better) than distributed policies, in which the content is eventually cached across the path from the requester node to the closest copy of the content. In some sense, we show that simplicity pays off.

The co-existence of cellular and wired networks has been exploited almost exclusively in the direction of OffLoading traffic from the former onto the latter. In this paper we claim that there exist cases that call for the exact opposite, i.e, use the cellular network to assist a fixed wired network. In particular, we show that by "OnLoading'' traffic from the wired broadband network onto the cellular network we can usefully speedup wired connections, on the downlink or the uplink. We consider the technological challenges pertaining to this idea and implement a prototype 3G OnLoading service that we call 3GOL, that can be deployed by an operator providing both the wired and cellular network services. By strategically OnLoading a fraction of the data transfers to the 3G network, one can significantly enhance the performance of particular applications. In particular we demonstrate non-trivial performance benefits of 3GOL to two widely used applications: video-on-demand and multimedia upload. We also consider the case when the operator that provides wired and cellular services is different, adding the analysis on economic constraints and volume cap on cellular data plans that need to be respected. Simulating 3GOL}over a DSLAM trace we show that 3GOL can reduce video pre-buffering time by at least 20% for 50% of the users while respecting data caps and we design a simple estimator to compute the daily allowance that can be used towards 3GOL while respecting caps. Our prototype is currently being piloted in 30 households in a large European city by a large network provider.

Cooperation among federated APs in dense urban areas can yield energy saving by allowing under-used devices to hand over their wireless stations (WS) to nearby APs and temporarily switch off while meeting user expectations in terms of throughput. We demonstrate the effectiveness and the benefits of our energy-efficient cooperative protocol through a real deployment emulating a residential scenario. The demo we propose is highly interactive, as users can generate traffic within a BSS through a wireless station, like a smartphone or a notebook, and observe, through a web interface, the protocol behavior and the network topology changes caused by the new traffic scenario.

The proliferation of overlapping, always-on IEEE 802.11 Access Points (APs) in urban areas can cause inefficient bandwidth usage and energy waste. Cooperation among federated APs could address these problems (i) by allowing under-used devices to hand over their wireless stations to nearby APs and temporarily switch off, (ii) by balancing the load of stations among APs and thus offloading congested APs. We outline a framework that allows such cooperation, yielding a 60% energy saving in realistic residential settings, while providing load balancing and meeting the user expectations in terms of throughput.

We present a lightweight algorithm for the estimation of the node achievable throughput and available bandwidth in IEEE 802.11 wireless networks. We consider a multirate WLAN with access point (AP), where there may be both elastic and inelastic traffic flows. Through our algorithm and leveraging previous theoretical results, the AP can estimate: (i) the available bandwidth that a new station wishing to associate with the AP can use, (ii) the impact on the system performance of admitting the new station, (iii) the bandwidth still available (if any) for inelastic traffic. The above quantities can be effectively used for admission control in WLANs and load balancing among APs with overlapping coverages. Indeed, simulation results show that the estimates yielded by our algorithm accurately reflect the system throughput behavior when there are both elastic and inelastic flows, in the uplink and downlink directions.

Multiple description coding (MDC) combined with multi-path transmission is a viable solution to the growing demand of reliable multimedia video communication over the Internet. Despite the progress in MDC techniques, transport protocols based on TCP cannot efficiently handle multipath transmissions of this kind of traffic. In this paper we present MD-SCTP, a full-multipath partially-reliable SCTP-based protocol, providing different scheduling schemes and featuring a selective retransmission within time-to-delivery constraints of multimedia traffic. Performance analysis proves the validity of our scheduler and of our selective retransmission scheme.