High resolution remote sensing for digital agriculture

Abstract

Sustainable agriculture in 21st century will face major challenges being mandatory to assimilate new technologies to increase agricultural productivity and sustainability. digital agriculture is a key technology to face those challenges that includes multiresolution remote sensing, accurate geodetic positioning, geographic information systems, and artificial intelligence. The main aim of this project is to study the use of multitemporal, multi-resolution, and multi-platform images collected from multiple sensors in various resolutions, various types of sensors: orbital, aerial high-resolution systems with UAVs, static ground and mobile will be used. The results will include the development of a terrestrial multi-sensor platform and the corresponding calibration processes, integrated orientation and generation of multispectral point clouds, which will be classified to generate information on the existence of pathogens, nutritional deficiencies, structural changes and fruit quantification or individual plants, being used in this project for Citrus, Coffee and other crops. The focus of this project will be the acquisition and data processing, integrating data coming from multiple sensors, to produce accurate and dense multitemporal multiband 3D data. This data will be analysed and classified with state of art machine learning techniques. Accuracy of multitemporal geospatial data will be a special concern as well as the use of artificial intelligence, intelligence that will support bioeconomic model for measure financial and production impact. In particular, this project will also strength cooperation with international players which will receive students and staff for one-year internships or short visits, deliver tutorials and post-graduate levels and promote joint researches. (AU)