Comparison of Unsupervised Algorithms for Vineyard Canopy Segmentation from UAV Multispectral Images

Abstract

Technical resources are currently supporting and enhancing the ability of precision agriculture techniques in crop management. The accuracy of prescription maps is a key aspect to ensure a fast and targeted intervention. In this context, remote sensing acquisition by unmanned aerial vehicles (UAV) is one of the most advanced platforms to collect imagery of the field. Besides the imagery acquisition, canopy segmentation among soil, plants and shadows is another practical and technical aspect that must be fast and precise to ensure a targeted intervention. In this paper, algorithms to be applied to UAV imagery are proposed according to the sensor used that could either be visible spectral or multispectral. These algorithms, called HSV-based (Hue, Saturation, Value), DEM (Digital Elevation Model) and K-means, are unsupervised, i.e., they perform canopy segmentation without human support. They were tested and compared in three different scenarios obtained from two vineyards over two years, 2017 and 2018 for RGB (Red-Green-Blue) and NRG (Near Infrared-Red-Green) imagery. Particular attention is given to the unsupervised ability of these algorithms to identify vines in these different acquisition conditions. This ability is quantified by the introduction of over- and under- estimation indexes, which are the algorithm’s ability to over-estimate or under-estimate vine canopies. For RGB imagery, the HSV-based algorithms consistently over-estimate vines, and never under-estimate them. The k-means and DEM method have a similar trend of under-estimation. While for NRG imagery, the HSV is the more stable algorithm and the DEM model slightly over-estimates the vines. HSV-based algorithms and the DEM algorithm have comparable computation time. The k-means algorithm increases computational demand as the quality of the DEM decreases. The algorithms developed can isolate canopy vegetation data, which is useful information about the current vineyard state, and can be used as a tool to be efficiently applied in the crop management procedure within precision viticulture applications.

Sentinel-2 Validation for Spatial Variability Assessment in Overhead Trellis System Viticulture Versus UAV and Agronomic Data

“…A radiometric calibration process was performed with MATLAB [90] to convert the digital number (DN) value for each pixel in radiance, by means of three OptoPolymer (OptoPolymer-Werner Sanftenberg, Munich, Germany) homogeneous and Lambertian surface panels (95%, 50%, and 5% reflectance). The filtering elaboration step to extract pure vine pixels was performed by applying an unsupervised algorithm developed in MATLAB and described in a previous paper by the authors [65]. This algorithm identifies soil, shadow, and vines by Otsu's global thresholding, using the HSV color spectrum.…”

“…In these cases, spatial analysis of remote sensing images requires separation of the canopy from bare soil, shadows, and green cover, which can be achieved with advanced filtering techniques for canopy extraction [56][57][58][59][60]. Many authors suggest unsupervised filtering techniques in a vineyard based on a canopy height model (CHM) derived from a digital elevation model (DEM) [51,53], geometric structure by texture analysis [61,62], 3D point cloud [63], automatic threshold on color distribution in RGB [64], different color spaces such as HSV (hue, saturation, value), or L*a*b* (L* for the lightness from black to white, a* from green to red and b* from blue to yellow) [65]. The work suggested by Cinat et al [65] presents an overview of those methodologies.…”

The Impact of Pan-Sharpening and Spectral Resolution on Vineyard Segmentation through Machine Learning

“…Analogous to this work, a number of other viticultural remote sensing studies have undertaken vine block detection using high resolution sub-metre pansharpened satellite imagery ( [4] using high pass filter (HPF) sharpening, and [18] with pan-sharpening method unknown). Other studies utilized higher resolution UAV or aerial imagery that does not require pansharpening for vinerow detection [6,12,[19][20][21][22][23] (other works using significantly coarser spatial resolutions will not be discussed in detail here, e.g., [8]). Detection completeness from these studies ranged from 72-90% of real vine blocks detected, with the main type of classification error being real vineyards that were missed/unclassified.…”

High-throughput phenotyping analysis of maize at the seedling stage using end-to-end segmentation network

“…At present, the threshold segmentation method is often used to segment top-view images of field crops. Although threshold segmentation with specific constraints can achieve very similar segmentation results [39,43], threshold segmentation is sensitive to noise and the effect on target segmentation is not ideal when there is little difference in gray scale. Threshold segmentation in different application scenarios (such as light and soil background) is relatively dependent on the selection of an empirical threshold.…”