Remote sensing is one of the key technologies for monitoring the environment and climate. On a global scale, open satellite data from the Landsat missions or ESA’s Copernicus program provide excellent information. However, the monitoring of local phenomena based on free satellite data, such as for vegetation, soil, or biodiversity observation, has been less successful so far due to the relatively large ground sampling rates (e.g., 10-60 m for Sentinel-2) that make the coverage of smaller agricultural areas inaccurate. In Austria, in particular, this is very important since about one-third of all agricultural areas are too small or too fragmented to be analyzed for crop types or harvest events using these satellite data. High error rates in automatic processing are the result.
Super-Resolution Reconstruction (SRR) offers a solution for monitoring these small-scale areas by enhancing image data with artificial intelligence. However, SRR technologies must first be further developed and optimized for practical application in agricultural monitoring to ensure higher resolution while maintaining the same spectral quality.
At the core of the SMAIL project lies the development of innovative solutions that enable the monitoring of small land parcels in remote sensing applications using publicly available satellite data. For this purpose, a customized task-based and AI-supported SRR application for time series of multispectral Sentinel-2 imagery is developed, allowing not only to enhance these images visually but also to solve tasks that were previously considered impossible with Sentinel-2 data. For example, this will support, among other things, the monitoring of cropping processes required for the transition to a more sustainable system of agricultural production in line with the European Green Deal. In addition, the derived vegetation and soil data can serve as the basis for more accurate environmental simulations for climate change adaptation.
Start: 05.2022
End Date: 04.2024
Agrarmarkt Austria (AMA), EOX IT Services GmbH (EOX), Paris-Lodron-Universität Salzburg, Department of Geoinformatics – Z_GIS
FFG - AI for Green
As early as 2023, the EU’s Common Agricultural Policy aims to replace time-consuming and costly on-site spot checks with comprehensive, digital monitoring of agricultural land. This opens up new opportunities in terms of operational and cost efficiency and, above all, for establishing state-of-the-art remote sensing applications in the data-driven agricultural management of the future. Here, the AI-based SRR solution of the SMAIL project, which can be integrated into existing agricultural monitoring workflows, makes an innovative contribution to reliable digital monitoring that supports professionals in reviewing and evaluating agricultural land.
Furthermore, the digital solutions envisioned in the SMAIL project are directly related to the UN Sustainable Development Goals (SDGs) and have a clear sustainable impact on the SDGs (15) Life on Land, (13) Climate Action, and (12) Responsible Consumption and Production. This results both from the use of AI technology, which enables more sustainable use of free satellite data, and from the digital solutions themselves, contributing to the optimization of comprehensive agricultural monitoring across Europe.
The goal of HORA 3D is the development and maintenance of a publicly accessible web application for object-related visualization of the results of the HORA 3 research project.
Development of extended reality technologies to create virtual maneuver scenarios based on real geographic data.
The main goal of this project is to enable a reliable decision support for large-scale infrastructure projects by providing solutions for a collaborative visual analysis of digital twins.
The research project CognitiveXR focuses on developing a platform that enables cognitive augmentation in the smart city domain by seamlessly integrating augmented reality, edge computing, and artificial intelligence.
For many years we have been dealing with all aspects of hydrodynamic modelling. With our software Visdom we can model different scenarios and also offer this as a service.
We create a digital twin of a community and simulate weather events to identify high-risk buildings and infrastructure vulnerabilities.
Visdom's hydrodynamic simulations provide precise water level predictions at any point on a water surface in the "PegelAlarm" app.
New technologies and security policy developments bring a paradigm shift for the Austrian armed forces. The use of information and communication measures entails both challenges and opportunities: for combat, security and rescue manoeuvres.
With the KAUST Scene Generator three-dimensional road networks can be generated from open-source OpenStreetMap data.
