Dual-branch vision network could support robotic pruning
A dual-branch network aims to segment grapevine stems and branches in images to support robotic pruning. The available material does not report its accuracy, field testing or readiness for commercial use.
- Source
- Google News – Vineyard Computer Vision
- Published
- Reading time
- 3 min read
- Region
- International

What the tool is about
The tool is described as a dual-branch network designed to segment grapevine stems and branches for robotic vineyard pruning. Segmentation means identifying which parts of an image belong to particular objects or plant structures.
This is a more demanding task than simply detecting whether a vine is present. Before a pruning robot can perform a physical operation, it needs to locate and outline the vine’s woody structure. According to the short material provided, this network focuses specifically on stems and branches.
How it works
The term “dual-branch” indicates that the network uses two processing branches. However, the available material does not explain what visual information each branch examines, how their outputs are combined or what kinds of vineyard images were used to train the system.
The likely output is a segmented image in which relevant vine parts are separated from the background and from one another. This is an application of computer vision, meaning the automated interpretation of camera images. The source excerpt provides no figures for accuracy or processing speed. It also gives no details about computing requirements, dataset size or comparisons with alternative methods.
Why it matters for vineyards and wineries
Reliable plant perception could form one part of an automated pruning workflow. If a machine can see stems and branches more precisely, it may gain better spatial information for later pruning decisions and for positioning a cutting tool. In the longer term, that could support more precise automation in the vineyard.
Segmentation alone, however, does not amount to a working pruning robot. A complete system would also have to select suitable cutting points, move its tool safely and cope with the variability found in commercial vineyards. The supplied material does not say how the network performs across different training systems, lighting conditions, visual obstructions, backgrounds or vine forms.
There is also no information about cost, commercial availability or integration with existing vineyard machinery. Growers and managers should therefore treat this as a technical signal rather than evidence of a market-ready product. No direct implications for cellar operations are described; the potential relevance is mainly in vineyard labour and automation.
What to watch next
The next useful signal would be the release of detailed technical results. Key questions include how accurately the network separates stems from branches, how often it makes errors and whether it can process images fast enough for real-time robotic operation.
Field testing will be especially important. A system that performs well on selected images may still face difficulties outdoors, where light, shadows, overlapping plant material and backgrounds can change continuously. It will also matter whether performance remains consistent across vineyards with different vine structures.
Ultimately, practical value will depend on whether the segmentation output can be connected to cutting-point planning, a robotic arm and a safe pruning mechanism. Based on the available information, it is not yet possible to say when or in what form the tool could reach vineyards.
The original publication named in the supplied material is Bioengineer.org.
Related articles
More stories from the same topic area.
Back to news
ResearchUV Robot Against Vineyard Mildew: What the 2026 Trial Could Show
During the 2026 growing season, a UV robot was trialled as a possible new tool for controlling vineyard mildew. The available material does not yet provide efficacy data, costs or detailed operational results.
Wine Industry Advisor · 11 November 2026 · 2 min read
TechnologyDual-branch network targets grapevine segmentation for robotic pruning
A dual-branch network could support robotic vineyard pruning by segmenting grapevine stems and branches. The available summary does not specify the model architecture, accuracy, training data or practical testing.
Google News – Vineyard Computer Vision · 13 November 2026 · 2 min read
TechnologyUV robot for fighting mildew joins Time’s best inventions of 2026
A UV robot designed to combat mildew has been named one of Time’s best inventions of 2026. Wine Industry Advisor trialled the machine throughout the 2026 growing season as a potential alternative to sulfur, although the available source material does not provide detailed results.
Wine Industry Advisor · 11 November 2026 · 2 min read