Search

The paper proposes a fully automatic algorithm approach to map burned areas from remote sensing characterized by human interpretable mapping criteria and explainable results. This approach is partially knowledge-driven and partially data-driven. It exploits active fire points to train the fusion function of factors deemed influential in determining the evidence of burned conditions from reflectance values of multispectral Sentinel-2 (S2) data. The fusion function is used to compute a map of seeds (burned pixels) that are adaptively expanded by applying a Region Growing (RG) algorithm to generate the final burned area map. The fusion function is an Ordered Weighted Averaging (OWA) operator, learnt through the application of a machine learning (ML) algorithm from a set of highly reliable fire points. Its semantics are characterized by two measures, the degrees of pessimism/optimism and democracy/monarchy. The former allows the prediction of the results of the fusion as affected by more false positives (commission errors) than false negatives (omission errors) in the case of pessimism, or vice versa; the latter foresees if there are only a few highly influential factors or many low influential ones that determine the result. The prediction on the degree of pessimism/optimism allows the expansion of the seeds to be appropriately tuned by selecting the most suited growing layer for the RG algorithm thus adapting the algorithm to the context. The paper illustrates the application of the automatic method in four study areas in southern Europe to map burned areas for the 2017 fire season. Thematic accuracy at each site was assessed by comparison to reference perimeters to prove the adaptability of the approach to the context; estimated average accuracy metrics are omission error = 0.057, commission error = 0.068, Dice coefficient = 0.94 and relative bias = 0.0046.

Sentinel-2 (S2) multi-spectral instrument (MSI) images are used in an automated approach built on fuzzy set theory and a region growing (RG) algorithm to identify areas affected by fires in Mediterranean regions. S2 spectral bands and their post- and pre-fire date (Δpost-pre) difference are interpreted as evidence of burn through soft constraints of membership functions defined from statistics of burned/unburned training regions; evidence of burn brought by the S2 spectral bands (partial evidence) is integrated using ordered weighted averaging (OWA) operators that provide synthetic score layers of likelihood of burn (global evidence of burn) that are combined in an RG algorithm. The algorithm is defined over a training site located in Italy, Vesuvius National Park, where membership functions are defined and OWA and RG algorithms are first tested. Over this site, validation is carried out by comparison with reference fire perimeters derived from supervised classification of very high-resolution (VHR) PlanetScope images leading to more than satisfactory results with Dice coefficient > 0.84, commission error < 0.22 and omission error < 0.15. The algorithm is tested for exportability over five sites in Portugal (1), Spain (2) and Greece (2) to evaluate the performance by comparison with fire reference perimeters derived from the Copernicus Emergency Management Service (EMS) database. In these sites, we estimate commission error < 0.15, omission error < 0.1 and Dice coefficient > 0.9 with accuracy in some cases greater than values obtained in the training site. Regression analysis confirmed the satisfactory accuracy levels achieved over all sites. The algorithm proposed offers the advantages of being least dependent on a priori/supervised selection for input bands (by building on the integration of redundant partial burn evidence) and for criteria/threshold to obtain segmentation into burned/unburned areas.

Machine learning can be used for social good. The employment of artificial intelligence in smart agriculture has many benefits for the environment: it helps small farmers (at a local scale) and policymakers and cooperatives (at regional scale) to take valid and coordinated countermeasures to combat climate change. This article discusses how artificial intelligence in agriculture can help to reduce costs, especially in developing countries such as Côte d’Ivoire, employing only low-cost or open-source tools, from hardware to software and open data. We developed machine learning models for two tasks: the first is improving agricultural farming cultivation, and the second is water management. For the first task, we used deep neural networks (YOLOv5m) to detect healthy plants and pods of cocoa and damaged ones only using mobile phone images. The results confirm it is possible to distinguish well the healthy from damaged ones. For actions at a larger scale, the second task proposes the analysis of remote sensors, coming from the GRACE NASA Mission and ERA5, produced by the Copernicus climate change service. A new deep neural network architecture (CIWA-net) is proposed with a U-Net-like architecture, aiming to forecast the total water storage anomalies. The model quality is compared to a vanilla convolutional neural network.

The challenge of this century is without a doubt to counter global warming. Land management, agriculture, and forests are responsible for 23% of total greenhouse gases (GHGs). In developing countries, such as those in African territories, where economic capacities are sometimes small and socio-cultural and linguistic perceptions are different, this requires a transition that is just and respectful of local culture and language, while at the same time helping to create labor conditions that respect gender and minors. This article describes a socio-technical ecological transition in the cocoa chain production in Côte d’Ivoire, which is not only the world’s leading producer of cocoa beans (45%) but also one of the African countries most prone to deforestation. Linguistic and multicultural aspects come together in Côte d’Ivoire, where more than 70 local languages are spoken. Intelligent digital approaches, agroecological issues, new methods, and sociocultural questions are embedded in a context of collaboration and co-creation, a living lab approach, to ensures sharing and co-creation among NGOs, farmers, companies, and researchers. A framework of socio-technical transition is presented, and this research aims to not only achieve the goals of a just ecological transition and reduce carbon footprint and deforestations but also to create diverse labour conditions, gender respect and inclusion.

Mycotoxigenic fungi can colonize small grain cereals causing severe yield losses and grain contaminations. Fusaria can be responsible for the contamination of wheat grains and derived products via several classes of mycotoxins, negatively impacting human and animal health. Among the strategies to control mycotoxins are analytical tools for their identification and quantification from field to food and feed. A fast multi-toxin assay based on reflective phantom interface (RPI) technology was developed to identify and quantify deoxynivalenol, zearalenone, and T-2/HT-2 toxins. The PRX analytical workflow was organized as follows: a fast mycotoxins extraction step followed by an analytical step carried out in a system composed of three elements: (I) a universal reader able to read a series of (II) cartridges that incorporate the RPI technology and (III) a software that analyzes data and gives feedback on the results. The assay was evaluated in wheat reference samples at known levels of toxin contaminations and on naturally contaminated grain samples. The results obtained suggest that the assay can be considered a useful screening tool for point-of-care and point-of-sale control of toxins contamination along wheat production and transformation chains.