Evaluation of novel precision viticulture tool for canopy biomass estimation and missing plant detection based on 2.5D and 3D approaches using RGB images acquired by UAV platform

Abstract

Background: The knowledge of vine vegetative status within a vineyard plays a key role in canopy management in order to achieve a correct vine balance and reach the final desired yield/quality. Detailed information about canopy architecture and missing plants distribution provides useful support for farmers/winegrowers to optimize canopy management practices and the replanting process, respectively. In the last decade, there has been a progressive diffusion of UAV (Unmanned Aerial Vehicles) technologies for Precision Viticulture purposes, as fast and accurate methodologies for spatial variability of geometric plant parameters. The aim of this study was to implement an unsupervised and integrated procedure of biomass estimation and missing plants detection, using both the 2.5D-surface and 3D-alphashape methods. Results: Both methods showed good overall accuracy respect to ground truth biomass measurements with high values of R 2 (0.71 and 0.80 for 2.5D and 3D, respectively). The 2.5D method led to an overestimation since it is derived by considering the vine as rectangular cuboid form. On the contrary, the 3D method provided more accurate results as a consequence of the alphashape algorithm, which is capable to detect each single shoot and holes within the canopy. Regarding the missing plants detection, the 3D approach confirmed better performance in cases of hidden conditions by shoots of adjacent plants or sparse canopy with some empty spaces along the row, where the 2.5D method based on the length of section of the row with lower thickness than the threshold used (0.10 m), tended to return false negatives and false positives, respectively. Conclusions: This paper describes a rapid and objective tool for the farmer to promptly identify canopy management strategies and drive replanting decisions. The 3D approach provided results closer to real canopy volume and higher performance in missing plant detection. However, the dense cloud based analysis required more processing time. In a future perspective, given the continuous technological evolution in terms of computing performance, the overcoming of the current limit represented by the pre-and post-processing phases of the large image dataset should mainstream this methodology.

Monitoring Vineyard Canopy Management Operations Using UAV-Acquired Photogrammetric Point Clouds

“…According to Campos et al [45], the combination of information extracted from UAV flights and a decision support system resulted in a 45% reduction of pesticide foliar application in vineyards. The 3D modelling of vineyards is also useful for the evaluation of vine-rows features (such as inter-row spacing or row orientation) [17] and for the detection of missing plants [46]. In addition, the estimation of vine dimensions throughout a season would allow the generation of dynamic growth maps, and the identification of areas were canopy management operations could be beneficial or needed to be re-designed [16].…”

Automatic Grapevine Trunk Detection on UAV-Based Point Cloud

“…The results are improved by selecting an early period (preferably belonging to phase 2- Figure 2) to conduct grapevine detection, with grapevines with a lower leaf cover, preventing the existence of vegetation in areas with no plants [21]. In Di Gennaro and Matese [44], 3D and 2.5D methods were compared for vineyard biomass and plant detection, but individual grapevine detection was not performed. In that study, as for missing plants, an overestimation was observed in the 3D-based method, false negatives were related to the existence of new plants while some false positives were due to different grapevine canopy thickness.…”

Beyond the traditional NDVI index as a key factor to mainstream the use of UAV in precision viticulture

“…Following this approach, it becomes possible to use high resolution multispectral images to individually quantify the components of the photosynthetically active biomass (PAB), obtaining the spectral response of the canopy from 2D analysis and biomass data from 3D analysis, thus overcoming the limit of the traditional NDVI index in the characterization of vigour. In support of the geometric approach is added the fact that it can be obtained with common RGB cameras, with minimum costs and very high performance derived from the extreme spatial resolution 6 . In general, the geometric approach presented in this paper can be applied on many tree crops, however on some exceptions with full soil covering such as overhead trellis system 21 and on extended crops such as horticultural crops or cereals, the NDVI index remains the only applicable solution for estimating production and vegetative parameters.…”

“…Geometric variables related to canopy thickness and volume were calculated using the 2.5D methodology with Matlab v.2019a (Mathworks, Natick, MA, USA) described in Di Gennaro and Matese (2020) 6 . Starting from the DEM of vines enclosed within a polygon grid, canopy thickness and vine number are extrapolated through the binarized image of the canopy extracted from each polygon grid.…”

“…In the last decade, the continuous advance in Unmanned Aerial Vehicle (UAV) technologies led to an increasing of spatial ground resolution 5 . Combining powerful photogrammetric software and image analysis methodologies it is possible to perform a geometric analysis at single vine level for canopy height, thickness and volume estimation 6 . As hypothesized by previous works 7 , 8 , this made it possible to identify new effective solutions to monitor vine vegetative and productive characteristics, with respect to traditional vegetation indices, such as the widely diffused Normalized Difference Vegetation Indices (NDVI).…”