MECS-VINE®: A New Proximal Sensor for Segmented Mapping of Vigor and Yield Parameters on Vineyard Rows

Abstract

Ground-based proximal sensing of vineyard features is gaining interest due to its ability to serve in even quite small plots with the advantage of being conducted concurrently with normal vineyard practices (i.e., spraying, pruning or soil tilling) with no dependence upon weather conditions, external services or law-imposed limitations. The purpose of the present work was to test performance of the new terrestrial multi-sensor MECS-VINE® in terms of reliability and degree of correlation with several canopy growth and yield parameters in the grapevine. MECS-VINE®, once conveniently positioned in front of the tractor, can provide simultaneous assessment of growth features and microclimate of specific canopy sections of the two adjacent row sides. MECS-VINE® integrates a series of microclimate sensors (air relative humidity, air and surface temperature) with two (left and right) matrix-based optical RGB imaging sensors and a related algorithm, termed Canoyct). MECS-VINE® was run five times along the season in a mature cv. Barbera vineyard and a Canopy Index (CI, pure number varying from 0 to 1000), calculated through its built-in algorithm, validated vs. canopy structure parameters (i.e., leaf layer number, fractions of canopy gaps and interior leaves) derived from point quadrat analysis. Results showed that CI was highly correlated vs. any canopy parameter at any date, although the closest relationships were found for CI vs. fraction of canopy gaps (R2 = 0.97) and leaf layer number (R2 = 0.97) for data pooled over 24 test vines. While correlations against canopy light interception and total lateral leaf area were still unsatisfactory, a good correlation was found vs. cluster and berry weight (R2 = 0.76 and 0.71, respectively) suggesting a good potential also for yield estimates. Besides the quite satisfactory calibration provided, main improvements of MECS-VINE® usage versus other current equipment are: (i) MECS-VINE® delivers a segmented evaluation of the canopy up to 15 different sectors, therefore allowing to differentiate canopy structure and density at specific and crucial canopy segments (i.e., basal part where clusters are located) and (ii) the sensor is optimized to work at any time of the day with any weather condition without the need of any supplemental lighting system.

On‐the‐go assessment of vineyard canopy porosity, bunch and leaf exposure by image analysis

“…Similarly, in Figure b, the proportion of exposed bunches by image analysis was 5.9% while a 36% value was obtained from PQA. Such differences may arise from the fact that in the PQA method the record of one insertion is extrapolated to an area of 100 cm 2 (if insertions are performed on a 10 × 10 cm grid, such as in Gatti et al ) to 240 cm 2 on average (in the present work). The inconsistencies in the extrapolation of each insertion record are larger for the minority elements, the bunches in Figure a,b.…”

“…Likewise, in Hill et al () and Diago et al () the image‐derived outcome was validated against the standard PQA, but a colour background and manual image acquisition were required. Improvements to on‐the‐go monitoring were recently described by Gatti et al (). These authors acquired RGB images on a stop‐and‐go mode using a conventional tractor (the tractor driver stopped and triggered the sensor), and obtained a significant correlation between an image‐derived canopy index value, which varied between 0 and 1000, and the fraction of canopy gaps and the leaf layer number, as measured by PQA.…”

“…The necessary steps towards automation and on‐the‐go canopy assessment using computer vision were outlined by Gatti et al (), who installed a system equipped with matrix‐based optical RGB sensors on a conventional tractor to assess, on a stop and go mode, the leaf layer number and fractions of canopy gaps and interior leaves during day time. Other authors have used RGB imagery acquired at night time from an autonomous moving platform, at low speed (0.3 km/h) to reconstruct a textured three‐dimensional (3D) point cloud to quantify yield components (Rose et al ).…”

On-the-Go Grapevine Yield Estimation Using Image Analysis and Boolean Model

“…Image analysis techniques allow for fast and reliable measurements, and recent studies have aimed its use in viticulture. Application examples include canopy status assessment [1,2] and, more recently, pruning mass determination [3]. As a noninvasive, reliable, and low-cost technology, image analysis is also a candidate for its integration in fully automated systems for vineyard monitoring [4].…”

Vineyard pruning weight assessment by machine vision: towards an on-the-go measurement system

“…In a similar experiment, Grocholsky et al (2011) measured the canopy shape and volume using a laser scanner and camera during the pre-harvest period obtaining an indirect estimation of the pruning weight within 10% of the actual value. Additionally, a new optical proximal sensor was developed in Italy for assessing and mapping the vigour and yield parameters in vineyards (Gatti et al, 2016). In Champagne (France), the Physiocap® sensor has appeared to be an interesting tool to assess vine vigour and to improve fertilization and pruning practices in precision viticulture (Debuisson et al, 2012;Demestihas et al, 2018).…”