Assessment of Vineyard Canopy Characteristics from Vigour Maps Obtained Using UAV and Satellite Imagery

Abstract

Canopy characterisation is a key factor for the success and efficiency of the pesticide application process in vineyards. Canopy measurements to determine the optimal volume rate are currently conducted manually, which is time-consuming and limits the adoption of precise methods for volume rate selection. Therefore, automated methods for canopy characterisation must be established using a rapid and reliable technology capable of providing precise information about crop structure. This research providedregression models for obtaining canopy characteristics of vineyards from unmanned aerial vehicle (UAV) and satellite images collected in three significant growth stages. Between 2018 and 2019, a total of 1400 vines were characterised manually and remotely using a UAV and a satellite-based technology. The information collected from the sampled vines was analysed by two different procedures. First, a linear relationship between the manual and remote sensing data was investigated considering every single vine as a data point. Second, the vines were clustered based on three vigour levels in the parcel, and regression models were fitted to the average values of the ground-based and remote sensing-estimated canopy parameters. Remote sensing could detect the changes in canopy characteristics associated with vegetation growth. The combination of normalised differential vegetation index (NDVI) and projected area extracted from the UAV images is correlated with the tree row volume (TRV) when raw point data were used. This relationship was improved and extended to canopy height, width, leaf wall area, and TRV when the data were clustered. Similarly, satellite-based NDVI yielded moderate coefficients of determination for canopy width with raw point data, and for canopy width, height, and TRV when the vines were clustered according to the vigour. The proposed approach should facilitate the estimation of canopy characteristics in each area of a field using a cost-effective, simple, and reliable technology, allowing variable rate application in vineyards.

Vineyard Pruning Weight Prediction Using 3D Point Clouds Generated from UAV Imagery and Structure from Motion Photogrammetry

“…Research exploring remote detection tools has led to the development of methods applying: laser image detection and ranging (LiDAR) to collect information on vine structure [7,9]; field spectroscopy to determine crop water content [10,11]; the MS Kinect device for pre-harvest production predictions [12]; and aerial photos from which to assess wine characteristics [13,14].…”

“…Such interest has been sparked by the shorter time needed to plan flights and replace sensors, along with the lower cost and high geometric resolution and precision timing afforded by such platforms [15][16][17][18]. Thanks to UAV imagery biomass can be estimated, missing plants identified [19][20][21], vigour maps drawn [13,22], and quality variables predicted [23,24] from vegetation indices and digital image processing.…”

Estimation of Leaf Area Index in vineyards by analysing projected shadows using UAV imagery

“…Some authors recommend MAE as the most natural measure of average error magnitude instead of RMSE, since measures of average error (such as RMSE), which are based on the sum of squared errors, are functions of the average error (MAE), the distribution of error magnitudes (or squared errors) and n 1/2 ; therefore, they do not describe average error alone and MAE is less sensitive to the effect of outliers than RMSE as an indicator of model performance (Willmott and Matsuura, 2005). However, RMSE is also indicated, since it is commonly used in Remote Sensing literature (López-Lozano et al, 2009;Li et al, 2014;Darvishzadeh et al, 2019;Beeri et al, 2020;Campos et al, 2021).…”

Integrating UAVs and Canopy Height Models in Vineyard Management: A Time-Space Approach

“…These estimations were extensively used to optimize canopy management and pest control, especially when variable rate technology (VRT) was employed [39]. More recently, the acquisition of high-resolution unmanned aerial vehicle (UAV) RGB imagery of the canopy has proved to be an effective tool for estimating plant architecture (e.g., vegetation height, canopy, density) through computation of accurate and reliable digital models [40][41][42][43]. The use of Ground Control Points (GCPs), located within the orchard's scene, represents an essential practice for spatial accuracy and minimization of model's errors [44,45].…”