06-13, 16:45–17:00 (Europe/London), Sala Videoconferenza @ PoliBa
Crop traits monitoring is a fundamental step for controlling crop productivity in the context of precision agriculture and field phenotyping. At present, the usage of hyperspectral data in machine learning regression algorithms (MLRAs) has attracted increasing attention to alleviate the challenges associated with traditional crop trait measurements. However, the performance assessment of such hyperspectral-based MLRA models for crop trait retrievals with respect to the well-known natural variations in either structural or biochemical crop properties remains largely elusive. As such, this experiment was set up to assess whether full-range hyperspectral data, acquired by a handheld spectrometer (Spectral Evolution; 350 – 2500 nm), as inputs to partial least squares regression (PLSR) and random forest (RF) models are capable of modeling different wheat crop traits at the canopy level. The examined crop traits were leaf area index (LAI), canopy water content (CWC), canopy chlorophyll content (CCC), and canopy nitrogen content (CNC). This approach allowed us, as an overarching objective, to compare the performance of the two aforementioned MLRA models while also focusing on the physical interpretation of the modelling results for each particular crop trait.
Overall, PLSR provided remarkably higher accuracy, tested with a cross-validation strategy, as compared to RF for all the crop traits. More precisely, PLSR denoted R2 (resp. nRMSE%) values of 0.72 (11.97), 0.77 (10.89), 0.70 (14.61), and 0.74 (14.38) for LAI, CWC, CCC, and CNC, respectively. All PLSR models indicated robust prediction capability with RPD values greater than 1.4, and amongst them, CWC was found to have excellent prediction performance with an RPD higher than 2. However, RF yielded less predictive models with R2 (resp. nRMSE%) values of 0.59 (14.59), 0.42 (17.42), 0.50 (18.86), and 0.42 (21.41) for LAI, CWC, CCC, and CNC, respectively. RF models for LAI and CCC showed good prediction capabilities (RPD > 1.4), whilst RF models of neither CWC nor CNC were reliable (RPD < 1.4).
In general, RF band importance and PLSR regression coefficient results revealed physically- meaningful and consistent patterns for each specific crop trait. Specific wavelengths at SWIR (1716-1745 nm) and NIR (1057-1120 nm), Green, and the Red-Edge bands respectively showed the highest importance for LAI retrieval. Water absorption regions around 910 nm and 1200 nm as well as the Red-Edge and Visible parts were of higher importance for the retrieval of CWC. The best-performing bands were situated in Red-Edge and Green spectral channels for CCC retrieval. SWIR spectral regions between 1600-1800 nm and 2100-2300 nm appeared to be important (in particular with respect to the other traits) alongside the Red-Edge part of the spectrum to retrieve CNC.
We demonstrated that full-range hyperspectral data in combination with MLRA algorithms can provide accurate estimates of wheat crop traits at the canopy level. The success of utilizing hyperspectral data in MLRA algorithms was further highlighted by the physically-meaningful modelling performances in accordance with the subtle structural and biochemical crop properties. Our results suggest that such spectroscopic hyperspectral-based MLRA approaches could be a powerful tool to accurately monitor crop status throughout the cropping season to improve high-throughput phenotyping activities and to further aid precision agricultural practices.
Mirco Boschetti received the Laurea Degree (summa cum laude) in Scienze Ambientali (1998) from the University of Milano Bicocca. He got the PhD from the State University of Milan (Agronomy faculty) with the thesis "Use of Remote Sensing and Crop Growth Model to monitor vegetation compound and cropping system". He has been with the National Research Council (CNR) of Italy since 1999 and he is a researcher at IREA since 2010. His research activity regards the use of remote sensing for vegetation and agro-ecosystem monitoring and the definition of environmental indicator through geographic multisource data integration. He has worked on automatic interpretation methods of multispectral and hyperspectral images acquired by airplane or satellite platform for the retrieval of bio-physical parameters and land cover mapping. In the last few years he is working on vegetation phenological parameters estimation, crop mapping and agro-practises monitoring from time series of low-resolution images. M.B. is member of the Scientific Board of the Italian Association of Remote Sensing (AIT). He is a reviewer of several Peer Review International Journal in the GIS, Remote Sensing, Environmental and Agronomical field. He participated at national (Ministry of Environment, Italian Space Agency) and international research programmes (EC, ESA). He was involved in the EU project "GEOLAND” and “GEOLAND2: towards a GMES Land Monitoring Core Service" (2008-2012, EU FP-7). Visiting scientist (in 2010) at IRRI head quarter (Los Banos, Philippines). Independent expert for the audit in the scope of the product and service independent evaluation of the Global land Component of the GIO Land service from 2015 -2017. He has been the coordinator of the FP7 SPACE project ERMES “An Earth obseRvation Model based RicE information Service”. Scientific responsible of the collaboration Framework between CNR-IREA and Bonifiche Ferraresi for research activities and technology transfer in ITC ad Remote Sensing for Precision Farming (2017 – 2020). WP leader of the project CHIME-RCS (2018 2019) and CCN (2019-2021) between ESA and ISPRA (Istituto Superiore per la Protezione e la Ricerca Ambientale) for the consolidation of requirement for Copernicus CHIME hyperspectral mission. Part of the CNR science team for the activities of CAL/VAL of PRISMA data for the Italian Space Agency (2019-2021). Since 2020 contract professor at University of Milan for the course “Geomatic: remote sensing for agriculture” (https://www.unimi.it/it/ugov/person/mirco-boschetti) .