This course has been conceived to teach the students how to process and analyze remotely sensed data acquired by UAVs for agricultural applications, with a focus on row crops. The students will learn about the components of integrated platforms, sensors, data that are acquired, and the analysis that is conducted for classical agricultural applications, including plant breeding and management practices.
Prerequisiti
Basic knowledge about physics (electromagnetic waves), remote sensing sensors (passive and active sensors) and data (digital numbers, processing levels).
Metodi didattici
Lectures and practicals using open access or commercial software.
Verifica Apprendimento
The students will be asked to answer a few questions about the topic covered by the course and show their ability o manage the software for a task assigned by the lecturer.
Testi
The slides used by the teacher will be provided, together with a list of relevant books and literature papers.
Contenuti
After an introductory lecture, the course will contain a series of lectures on: 1. UAV system description: Platforms configurations and sensor technologies 2. Current state-of-the-art in precision agriculture using UAVs 3. Basics of data acquisition (including ground reference data) and processing for remote sensing applications (spectral and geometric calibration/correction) 4. Utilizing multispectral, hyperspectral, and LiDAR sensors for applications in agriculture, including phenomics for plant breeding and management practice optimization 5. Traditional and state-of-the-art methodology for analysis of remotely sensed data for agricultural applications (detection and counting, classification, prediction). 6. Generalization of models over space and time 7. Looking forward: Multi-modality, multi-resolution remote sensing for agricultural applications A few case studies will be provided: P1: The role of field reference and environmental data in agricultural applications of remote sensing P2: Developing high resolution RGB orthomosaics P3: LiDAR data analysis for plant heights plant structure P4: Hyperspectral data analysis in analysis of plant traits P5: Prediction of yield using classical ML and multi-temporal deep learning approaches.
