A computer vision system for early stage grape yield estimation based on shoot detection

Liu, Scarlett; Cossell, Steve; Tang, Julie; Dunn, Gregory; Whitty, Mark

doi:10.1016/j.compag.2017.03.013

Cited by 71 publications

(49 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Computer vision is a non‐invasive technology, which involves the automated acquisition, analysis and understanding of useful information from a single image or a sequence of images. In viticulture, the use of computer vision outdoors from still photography with visible [red green blue (RGB)] cameras has been used to characterise different features of the vineyard (Tardaguila et al , Hill et al ) and to estimate yield components (Dunn and Martin , Nuske et al , Liu et al ). Recent work has advanced our ability to assess canopy features from RGB imaging in grapevines (Diago et al , ) and in other crops (Chopin et al , ) by adopting colour corrections and hybrid approaches of the classification algorithms.…”

Section: Introductionmentioning

confidence: 99%

“…Recent work has advanced our ability to assess canopy features from RGB imaging in grapevines (Diago et al , ) and in other crops (Chopin et al , ) by adopting colour corrections and hybrid approaches of the classification algorithms. With the exception of the work of Nuske et al () and Liu et al (), studies have involved static, point‐to‐point, manual image acquisition, which from a practical standpoint, is a constraint when a large number of grapevines need to be assessed. Thus, there are still some practical limitations to the adoption of RGB imaging as a commercial monitoring method, such as the manual and static mode of image acquisition or the use of a colour background.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Diago

Aquino

Millán

et al. 2019

Australian Journal of Grape and Wine Research

View full text Add to dashboard Cite

Background and Aims Canopy assessment of the fruiting zone can lead to more informed vineyard management decisions. A non‐destructive, image‐based system capable of operating on‐the‐go was developed to assess canopy porosity, and leaf and bunch exposure of red grape cultivars in the vineyard. Methods and Results On‐the‐go (7 km/h) night time images of a vertically shoot positioned commercial vineyard canopy were acquired with an automated red green blue imaging system, coupled to a GPS and controlled artificial lighting. The reference method was point quadrat analysis. Sound correlations between the image analysis and point quadrat analysis results for the proportion of gaps (R2 > 0.85; P < 0.001) and leaf to canopy area ratio (R2 > 0.57; P < 0.001) were obtained for both sides of the canopy. For the bunch to canopy area ratio the best relationship was found on the western side of the canopy (R2 = 0.79; P < 0.001). Also maps of the three canopy variables were built in a commercial vineyard to compare their spatial variability on the east and west sides across the whole vineyard plot. Conclusions The developed imaging system, capable of operating on‐the‐go, can yield quantitative, objective and reliable knowledge of what a grapegrower would assess by subjective, qualitative visual inspection of the grapevine canopy. The information can be used to help make better informed decisions about leaf removal, and if mapped may help to delineate zones amenable to homogeneous management. Significance of the Study The new developed computer vision system can be mounted on any vehicle, such as a tractor, all terrain vehicle and robot, for a rapid and objective monitoring of the vineyard canopy around the fruiting zone in red cultivars and vertically shoot positioned trained vines. Moreover, the maps generated could be used by a new generation of variable rate viticultural machinery to spatially optimise vineyard cultural practices.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

Diago

Aquino

Millán

et al. 2019

Australian Journal of Grape and Wine Research

View full text Add to dashboard Cite

show abstract

“…Recent progress in optically based sensor approaches to yield estimation [e.g. Nuske et al (), Diago et al (), Liu et al ()] may provide an alternative and simpler approach to early season yield estimation. Problems posed due to bunch occlusion by leaves and other bunches presents a significant challenge to bunch number/mass estimation once the canopy has developed post‐flowering.…”

Section: Resultsmentioning

confidence: 99%

“…Seasonal differences in grape yield present significant challenges to both grapegrowers and winemakers (Trought ), which is a major reason for the recent research focus placed on yield estimation (Dunn and Martin , Diago et al , , Nuske et al , Herrero‐Huerta et al , Liu et al ). Such work is premised on the idea that understanding the sources of variation in yield and yield components will help improve the methods and accuracy of pre‐harvest yield estimates.…”

Section: Introductionmentioning

confidence: 99%

Spatio‐temporal variability in vine vigour and yield in a Marlborough Sauvignon Blanc vineyard

Bramley

Ouzman

Trought

et al. 2019

Australian Journal of Grape and Wine Research

View full text Add to dashboard Cite

Background and Aims Previous work has indicated that patterns of within‐vineyard variation in vine vigour are stable in time and, in the case of spur‐pruned vineyards, closely match patterns of variation in yield. However, whether this yield : vigour interaction also occurs in cane‐pruned vineyards is uncertain. This work sought to better understand this issue in support of efforts to improve yield estimation in cane‐pruned Marlborough Sauvignon Blanc. Methods and Results Vine vigour, measured as pruning mass, trunk cross‐sectional area and using remotely sensed imagery, and the components of yield were measured in a 5.9 ha Marlborough Sauvignon Blanc vineyard on vines which had been pruned to retain either two or four canes. The results suggest that whereas patterns of variation in vine vigour are stable in time, and related to variation in the land underlying the vineyard, patterns of yield variation are neither temporally stable nor related to variation in vine vigour or the inherent underlying characteristics of the block. Conclusions For all practical purposes, variation in the yield of Marlborough Sauvignon Blanc within vineyard ‘zones’ identified on the basis of variation in vine vigour or soil properties can be regarded as random. Nevertheless, given a diversity of product offering, and because variation in vigour can have a marked impact on fruit composition, remotely sensed indices of vigour do provide a valuable basis for targeting sampling or sensor deployment aimed at yield estimation. Significance of the Study This work highlights the need to differentiate between spur‐ and cane‐pruned vineyards in considering how vineyard variability should be assessed, and how the tools of precision viticulture and targeted management are applied.

show abstract

“…Alternatively, in-field imaging does not require destructive sample collection, thus increasing its industrial applicability, with use cases including the estimation of the number of flowers per inflorescence (Millan et al, 2017;Liu et al, 2018), the assessment of canopy architecture (Diago et al, 2016a) or grapevine phenotyping (Kicherer et al, 2015;Klodt et al, 2015). Image sensors have been mounted on agricultural vehicles for yield prediction (Nuske et al, 2011;Liu et al, 2017;Aquino et al, 2018;.…”

Section: Introductionmentioning

confidence: 99%

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

et al. 2019

View full text Add to dashboard Cite

Aim: Pruning weight is an indicator of vegetative growth and vigour in grapevine. Traditionally, it is manually determined, which is time-consuming and labour-demanding. This study aims at providing a new, non-invasive and low-cost method for pruning weight estimation in commercial vineyards based on computer vision.Methods and results: The methodology relies on computer-based analysis of RGB images captured manually and on-the-go in a VSP Tempranillo vineyard. Firstly, the pruning weight estimation was evaluated using manually taken photographs using a controlled background. These images were analysed to generate a model of wood pruning weight estimation, resulting in a coefficient of determination (R2) of 0.91 (p<0.001) and a root-mean-square error (RMSE) of 87.7 g. After this, a mobile sensor platform (modified ATV) was used to take vine images automatically and on-the-go without background. These RGB images were analysed using a fully automated computer vision algorithm, resulting in R2 = 0.75 (p<0.001) and RMSE = 147.9 g. Finally, the mobile sensor platform was also used to sample a commercial VSP vineyard to map the spatial variability of wood pruning weight, and hereafter vine vigour.Conclusions: The results showed that the developed computer vision methodology was able to estimate the vine pruning weight in commercial vineyards and to map the spatial variation of the pruning weight across a vineyard.Significance and impact of the study: The presented methodology may become a valuable tool for the wine industry for rapid assessment and mapping of vine vigour. This information can be used to support decision making on pruning, fertilization and canopy management.

show abstract

A computer vision system for early stage grape yield estimation based on shoot detection

Cited by 71 publications

References 11 publications

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

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

Spatio‐temporal variability in vine vigour and yield in a Marlborough Sauvignon Blanc vineyard

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

Contact Info

Product

Resources

About