The growing amount of data collected by Earth Observation (EO) satellites requires new processing procedures able to manage huge quantity of information. Artificial Intelligence and Deep Learning (DL) can provide advanced information also because of their ability to extract valuable information from complex data. Thanks to specific hardware platforms, these algorithms can be used also in space, opening the possibility for new procedures for intelligent data processing. The European Space Agency Φ-Sat-2 mission was designed with the purpose of demonstrating the benefits of using AI in space by running AI-based applications on board a CubeSat. We present here the Convolutional AutoEncoder-based algorithm developed for on-board lossy image compression of the Φ-Sat-2 mission and provide a first benchmark addressing a real space mission and a new image compression end-to-end architecture based on AI. Image compression is a crucial application that allows to save transmission bandwidth and storage. In fact, images acquired by the sensor can be compressed on-board and sent to the ground where they are reconstructed. DL algorithms have already been successfully applied for image compression however performance degradation may occur in the context of a representative on-board environment. Therefore, besides analyzing the results for the local hardware environment, this paper investigates the performance variation for the on-board setting. An additional piece of innovation is the introduction of an applicative metric for the evaluation of the compression to assess the applicability of the reconstructed images for other tasks. Such metric completes those more traditional based on the original-reconstructed image similarity.

Remote sensing can represent an important instrument for monitoring landfills and their evolution over time. In general, remote sensing can offer a global and rapid view of the Earth’s surface. Thanks to a wide variety of heterogeneous sensors, it can provide high-level information, making it a useful technology for many applications. The main purpose of this paper is to provide a review of relevant methods based on remote sensing for landfill identification and monitoring. The methods found in the literature make use of measurements acquired from both multi-spectral and radar sensors and exploit vegetation indexes, land surface temperature, and backscatter information, either separately or in combination. Moreover, additional information can be provided by atmospheric sounders able to detect gas emissions (e.g., methane) and hyperspectral sensors. In order to provide a comprehensive overview of the full potential of Earth observation data for landfill monitoring, this article also provides applications of the main procedures presented to selected test sites. These applications highlight the potentialities of satellite-borne sensors for improving the detection and delimitation of landfills and enhancing the evaluation of waste disposal effects on environmental health. The results revealed that a single-sensor-based analysis can provide significant information on the landfill evolution. However, a data fusion approach that incorporates data acquired from heterogeneous sensors, including visible/near infrared, thermal infrared, and synthetic aperture radar (SAR), can result in a more effective instrument to fully support the monitoring of landfills and their effect on the surrounding area. In particular, the results show that a synergistic use of multispectral indexes, land surface temperature, and the backscatter coefficient retrieved from SAR sensors can improve the sensitivity to changes in the spatial geometry of the considered site.

The general consensus on future climate projections poses new and increased concerns about climate change and its impacts. Droughts are primarily worrying, since they contribute to altering the composition, distribution, and abundance of species. Grasslands, for example, are the primary source for grazing mammals and modifications in climate determine variation in the available yields for cattle. To support the agriculture sector, international organizations such as the Food and Agriculture Organization (FAO) of the United Nations are promoting the development of dedicated monitoring initiatives, with particular attention for undeveloped and disadvantaged countries. The temporal scale is very important in this context, where long time series of data are required to compute consistent analyses. In this research, we discuss the results regarding long-term grass biomass estimation in an extended African region. The results are obtained by means of a procedure that is mostly automatic and replicable in other contexts. Zambia has been identified as a significant test area due to its vulnerability to the adverse impacts of climate change as a result of its geographic location, socioeconomic stresses, and low adaptive capacity. In fact, analysis and estimations were performed over a long time window (21 years) to identify correlations with climate variables, such as precipitation, to clarify sensitivity to climate change and possible effects already in place. From the analysis, decline in both grass quality and quantity was not currently evident in the study area. However, pastures in the considered area were found to be vulnerable to changing climate and, in particular, to the water shortages accompanying drought periods.

Studies of oil slicks in the ocean environment with synthetic aperture radar (SAR) have found that one of the most complex challenges to oil spill detection is the separation of mineral oil spills from slicks that are biogenic in origin. The possible occurrence of multiple scattering mechanisms beyond Bragg scattering for the sea surface, with or without biogenic or mineral oil slicks, and even under low to moderate wind conditions, has also been a subject of debate because the measured signals from these radar-dark surfaces can be contaminated easily by noise. Therefore, the use of noise-uncontaminated data is required for oil spill study in order to avoid significant alteration in the measured radar backscatter, which can lead to misinterpretation and misclassification of the scattering mechanisms involved. To this end, this study uses uninhabited aerial vehicle SAR data, with a noise-equivalent sigma zero as low as −53 dB, to investigate slick classification within a deep learning framework in order to assess deep architectures’ capabilities for providing a reliable and accurate three-state classifier capable of separating mineral oil films from biogenic slicks and from the clean sea. The study exploits parameters with sensitivity to the dielectric constant and ocean wave damping properties, and convolutional neural networks’ (CNNs’) capability for learning nonlinear features, shapes, and textural and statistical patterns, in order to obtain significant classification accuracy. Very high accuracy results have been achieved, with values up to 0.91, 0.94, 0.98, and 0.99 under the most probable real-world spill acquisition conditions.

Evaluation of the impact of climate change on water bodies has been one of the most discussed open issues of recent years. The exploitation of satellite data for the monitoring of water surface temperatures, combined with ground measurements where available, has already been shown in several previous studies, but these studies mainly focused on large lakes around the world. In this work the water surface temperature characterization during the last few decades of two small–medium Italian lakes, Lake Bracciano and Lake Martignano, using satellite data is addressed. The study also takes advantage of the last space-borne platforms, such as Sentinel-3. Long time series of clear sky conditions and atmospherically calibrated (using a simplified Planck’s Law-based algorithm) images were processed in order to derive the lakes surface temperature trends from 1984 to 2019. The results show an overall increase in water surface temperatures which is more evident on the smallest and shallowest of the two test sites. In particular, it was observed that, since the year 2000, the surface temperature of both lakes has risen by about 0.106 °C/year on average, which doubles the rate that can be retrieved by considering the whole period 1984–2019 (0.053 °C/year on average).

The increase in remote sensing satellite imagery with high spatial and temporal resolutions has enabled the development of a wide variety of applications for Earth observation and monitoring. At the same time, it requires new techniques that are able to manage the amount of data stored and transmitted to the ground. Advanced techniques for on-board data processing answer this problem, offering the possibility to select only the data of interest for a specific application or to extract specific information from data. However, the computational resources that exist on-board are limited compared to the ground segment availability. Alternatively, in applications such as change detection, only images containing changes are useful and worth being stored and sent to the ground. In this paper, we propose a change detection scheme that could be run on-board. It relies on a feature-based representation of the acquired images which is obtained by means of an auto-associative neural network (AANN). Once the AANN is trained, the dissimilarity between two images is evaluated in terms of the extracted features. This information can be subsequently turned into a change detection result. This study, which presents one of the first techniques for on-board change detection, yielded encouraging results on a set of Sentinel-2 images, even in light of comparison with a benchmark technique.