Addressing crop-mapping problems using high-resolution hyperspectral images requires innovative solutions that integrate advanced image processing and machine learning. In this paper, we propose an approach designed to address the main challenges of crop mapping using hyperspectral images, including the analysis of high-dimensional feature spaces, the limited number of training samples and the complex spectral-spatial relationships. The proposed architecture exploits Contractive-Expansive-Contractive (CEC) connections, which extract features capable of capturing both global and local patterns, while enhance discriminative features and suppress noise and irrelevant information. The CEC network is used to integrate multiscale features and learn the optimal decision boundaries for segmentation. The proposed architecture is evaluated on a Prisma hyperspectral dataset and compared against other methodologies suited for crop-type mapping. The experimental results demonstrate that the proposed approach achieves higher overall and class-wise segmentation accuracy.

Artificial Intelligence (AI) represents a collection of tools and methodologies that have the potential to revolutionise various aspects of human activity. Earth observation (EO) data, including satellite and in-situ, are essential in a number of high impact applications, ranging from security and energy to agriculture and health. In this paper, we present the AI4Copernicus framework for bridging the two domains within the European context to enable data-centred innovation. In order to achieve this goal, AI4Copernicus has developed and enriches the European AI-on-demand platform with a number of application bootstrapping services and tools to accelerate uptake and innovation, whilst it provides integration over AI-on-Demand services and the Copernicus ecosystem, targeting the highly successful Data and Information Access Service (DIAS) Cloud platforms. More specifically, by employing procedures for onboarding and validating models and tools, and by utilising a host of meticulously reviewed and supervised open calls-enabled projects, and containerisation best-practices, AI4Copernicus deployed and made available several products on DIAS platforms. Moreover, these products and resources have been made available on the AI-on-Demand platform catalogue for discovery, use and further development. The AI4Copernicus framework is being used by a number of business-driven projects and SMEs spanning several application domains. This article provides an overview of the European AI and EO context as well as the AI4Copernicus technological framework and tools offered. Further, we present real world use-cases as well as a community-centred evaluation of our framework based on usage and feedback received from several projects.

The temporal consistency of yearly land-cover maps is of great importance to model the evolution and change of the land cover over the years. In this paper, we focus the attention on a novel approach to classification of yearly satellite image time series (SITS) that combines deep learning with Bayesian modelling, using Hidden Markov Models (HMMs) integrated with Transformer Encoder (TE) based DNNs. The proposed approach aims to capture both i) intricate temporal correlations in yearly SITS and ii) specific patterns in multiyear crop type sequences. It leverages the cascade classification of an HMM layer built on top of the TE, discerning consistent yearly crop-type sequences. Validation on a multiyear crop type classification dataset spanning 47 crop types and six years of Sentinel-2 acquisitions demonstrates the importance of modelling temporal consistency in the predicted labels. HMMs enhance the overall performance and F1 scores, emphasising the effectiveness of the proposed approach.

This article presents a novel system that produces multiyear high-resolution irrigation water demand maps for agricultural areas, enabling a new level of detail for irrigation support for farmers and agricultural stakeholders. The system is based on a scalable distributed deep learning (DL) model trained on dense time series of Sentinel-2 images and a large training set for the first year of observation and fine tuned on new labeled data for the consecutive years. The trained models are used to generate multiyear crop type maps, which are assimilated together with the Sentinel-2 dense time series and the meteorological data into a physically based agrohydrological model to derive the irrigation water demand for different crops. To process the required large volume of multiyear Copernicus Sentinel-2 data, the software architecture of the proposed system has been built on the integration of the Food Security thematic exploitation platform (TEP) and the data-intensive artificial intelligence Hopsworks platform. While the Food Security TEP provides easy access to Sentinel-2 data and the possibility of developing processing algorithms directly in the cloud, the Hopsworks platform has been used to train DL algorithms in a distributed manner. The experimental analysis was carried out in the upper part of the Danube Basin for the years 2018, 2019, and 2020 considering 37 Sentinel-2 tiles acquired in Austria, Moravia, Hungary, Slovakia, and Germany.

This work presents a multitemporal class-driven hierarchical Residual Neural Network (ResNet) designed for modelling the classification of Time Series (TS) of multispectral images at different semantical class levels. The architecture consists of a modification of the ResNet where we introduce additional branches to perform the classification at the different hierarchy levels and leverage on hierarchy-penalty maps to discourage incoherent hierarchical transitions within the classification. In this way, we improve the discrimination capabilities of classes at different levels of semantic details and train a modular architecture that can be used as a backbone network for introducing new specific classes and additional tasks considering limited training samples available. We exploit the class-hierarchy labels to train efficiently the different layers of the architecture, allowing the first layers to train faster on the first levels of the hierarchy modeling general classes (i.e., the macro-classes) and the intermediate classes, while using the last ones to discriminate more specific classes (i.e., the micro-classes). In this way, the targets are constrained in following the hierarchy defined, improving the classification of classes at the most detailed level. The proposed modular network has intrinsic adaptation capability that can be obtained through fine tuning. The experimental results, obtained on two tiles of the Amazonian Forest on 12 monthly composites of Sentinel-2 images acquired during 2019, demonstrate the effectiveness of the hierarchical approach in both generalizing over different hierarchical levels and learning discriminant features for an accurate classification at the micro-class level on a new target area, with a better representation of the minoritarian classes.

Artificial Intelligence (AI) represents a collection of tools and methodologies that have the potential of transforming virtually all aspects of human activity. Earth observation (EO) data, including satellite and in-situ, are essential for a number of applications, covering high-impact domains as diverse as security, agriculture, energy and health. In this paper, we present the AI4Copernicus framework for bridging the two domains within the European context to enable data-centred innovation. In order to achieve this goal, AI4Copernicus enriches the European AI-on-demand platform with a number of bootstrapping services and tools to accelerate uptake and innovation, whilst it provides integration over AI-on-Demand services and the Copernicus ecosystem over the highly successful Data and Information Access Services (DIAS) Cloud platforms. The AI4Copernicus framework is being used by a number of business-driven projects spanning several application domains. In this paper, we provide an overview of the European AI and EO approach as well as of the AI4Copernicus technological framework and tools offered. Further, we describe exemplary real world use-cases as well as technological evaluation of our framework based on usage and feedback received from a number of projects.

Food security, especially in a changing Earth environment, is one of the most challenging issues of this century. Population growth, increased food consumption and the challenges of climate change will extend over the next decades. To deal with these, both regional and global measures are necessary. Biomass production and thus yield will need to be increased in a sustainable way. It is important to minimize the risks of yield loss even under more extreme environmental conditions, while making sure not to deplete or damage the available resources. Two measures are most important for this: irrigation and fertilization. While fertilization relies mainly on industrial goods, irrigation requires reliable water resources in the area that is being farmed, either from groundwater or surface water. Regarding surface water, a large portion of the world’s fresh-water is linked to snowfall, snow storage and seasonal release of the water. All these components are subject to increased variability due to climate change and the

The Danube Basin has been hit by several droughts in the last few years. As climate change makes weather extremes and temperature records in late winter and early spring more likely, water availability and irrigation possibilities become more important. In this paper, the crop water demand at field and national scale within the Danube Basin is presented using a dense time series of multispectral Sentinel-2 data, for crop type maps derived with deep learning techniques and physically-based models for crop parameter retrieval and crop growth modelling.

ExtremeEarth is a three-year H2020 ICT research and innovation project. Its main objective is to develop Artificial Intelligence and big data technologies that scale to the large volumes of big Copernicus data, information and knowledge, and apply these technologies in two of the European Space Agency (ESA) Thematic Exploitation Platforms (TEP): Food Security and Polar.

Bringing together a number of cutting-edge technologies that range from storing extremely large volumes of data all the way to developing scalable machine learning and deep learning algorithms in a distributed manner and having them operate over the same infrastructure poses unprecedented challenges. One of these challenges is the integration of European Space Agency (ESA)'s Thematic Exploitation Platforms (TEPs) and data information access service platforms with a data platform, namely Hopsworks, which enables scalable data processing, machine learning, and deep learning on Copernicus data, and development of very large training datasets for deep learning architectures targeting the classification of Sentinel images. In this article, we present the software architecture of ExtremeEarth that aims at the development of scalable deep learning and geospatial analytics techniques for processing and analyzing petabytes of Copernicus data. The ExtremeEarth software infrastructure seamlessly integrates existing and novel software platforms and tools for storing, accessing, processing, analyzing, and visualizing large amounts of Copernicus data. New techniques in the areas of remote sensing and artificial intelligence with an emphasis on deep learning are developed. These techniques and corresponding software presented in this article are to be integrated with and used in two ESA TEPs, namely Polar and Food Security TEPs. Furthermore, we present the integration of Hopsworks with the Polar and Food Security use cases and the flow of events for the products offered through the TEPs.

This work presents a system for multi-year crop type mapping based on the multi-temporal Long Short-Term Memory (LSTM) Deep Learning (DL) model and Sentinel 2 image Time Series (TS). The method assumes the availability of a pre-trained LSTM model for a given year and aims to update the corresponding crop type map fora different year considering a small amount of recent reference data. To this end, the proposed approach combines Self-Paced Learning (SPL) and fine-tuning (FT) techniques. While the SPL technique gradually incorporates samples from crop types that can be classified with high-confidence by the pre-trained model, the FT strategy adapts the network to those classes having low-confidence accuracy. This condition allows us to reduce the labeled samples required to achieve accurate classification results. The experimental results obtained on three tiles of the Austrian country on TSs of Sentinel 2 data acquired in 2019 and 2020 (considering a model pre-trained on images of 2018) demonstrate the capability of the LSTM to adapt to TS of images with different temporal and radiometric characteristic with respect to the one used to pre-train the model, with a relatively small number of training samples. As expected, by directly applying the model without performing any adaptation, we obtain a mean F-score (F1%) of 64% and 62% compared to 76% and 70% achieved by the proposed technique with only 1500 samples for 2019 and 2020, respectively.

This article presents TimeSen2Crop, a pixel-based dataset made up of more than 1 million samples of Sentinel 2 time series (TSs) associated to 16 crop types. This dataset, publicly available, aims to contribute to the worldwide research related to the supervised classification of TSs of Sentinel 2 data for crop type mapping. TimeSen2Crop includes atmospherically corrected images and reports the snow, shadows, and clouds information per labeled unit. The provided TSs represent an agronomic year (i.e., period from one year's harvest to the next one for agricultural commodity) ranging from September 2017 to August 2018. To generate the dataset, the publicly available Austrian crop type map based on farmer's declarations has been considered. To ensure the selection of reliable labeled units from the map (i.e., pure pixels correctly associated to their labels), an automatic procedure for the extraction of the training set based on a multitemporal deep learning model has been defined. TimeSen2Crop also includes a TS of Sentinel 2 images acquired in the following agronomic year (i.e., from September 2018 to August 2019). These data are provided with the aim of attract more research activities for solving a typical challenge of the crop type mapping task: adapting multitemporal deep learning models to different year (domain adaptation). The design of the dataset is described along with a benchmark comparison of deep learning models for crop type mapping.

The regular monitoring of agricultural areas is extremely important for mitigating food insecurity risks and for planning government interventions. In the literature, several deep learning algorithms have been recently proposed to perform land cover/ land use classification by using multispectral optical images. However, most of the considered deep learning models, such as the standard Convolutional Neural Networks (CNN), rely on mono-temporal images, focusing on spectral and textural features while discarding the temporal component, which is crucial for the accurate crop type mapping. In this work, we exploit a Long Short Term Memory (LSTM) deep learning classification architecture to characterize agricultural area dynamics by using the multitemporal multispectral information provided by satellite multispectral sensor Sentinel 2. Instead of considering a pre-trained network and applying to it a fine-tuning, the proposed architecture is trained from scratch in order to be tailored to the specific properties of the long time series of Sentinel 2 multispectral images. To face the lack of labeled training database, existing crop type maps available at the country level are used to generate a large set of weak reference data. First, the proposed method automatically extracts a large training dataset from existing crop type maps, by detecting those samples having the highest probability of being correctly classified. Then, the weak labeled samples extracted are used to train the deep LSTM architecture on a time series of Sentinel 2 images acquired over an entire year. The preliminary results obtained demonstrate the effectiveness of the proposed approach, which is promising at large scale.