Forest surveillance with artificial intelligence and digital technologies - FORSAID

Forests have an important role in the achievement of the objectives of the European Green Deal. Forest trees are increasingly threatened by invasive pests, with many of them being regulated in the Union territory.

FORSAID has as an overall goal to develop a comprehensive combination of innovative digital technologies aimed at detecting regulated forest pests at an early stage, surveying their occurrence in the territory, and providing essential information for the adoption of phytosanitary measures to limit their spread and impacts.

The project adopts a multi-actor and multidisciplinary approach tailored to develop and favour the adoption of digital technologies at different spatial and temporal scales associated with a selected list of important regulated forest pests.

The Internet of Things (IoT) will be used to create networks of insect traps for major guilds of insects, thanks to innovative deep learning analysis of images sent remotely from the traps. Robotized devices will be developed and tested for the automatic barcoding of the captured pests.

Drones equipped with sensors will be used for the scanning of plant health status through the measure of physiological variables.

Remote sensing techniques will be used to validate existing ground-truth data on the occurrence of tree alterations associated with abiotic and biotic factors, and models based on Artificial Intelligence (AI) and machine learning (ML) will be developed to discriminate different types of stresses as soon as they appear.

An economic analysis will address the costs and benefits of using digital technologies for the detection and surveillance measures, considering the economic, environmental, and social impacts of regulated pests in EU forests.

Stakeholders from the forest sector will be involved in a multi-actor approach to drive the research to applicable results and co-construct guidelines for the best use of new digital technologies for forest pest detection and monitoring.