PhD position: Optimizing images quality and deep learning methods vineyard disease detection

Position: PhD position: Optimizing images quality and deep learning methods for vineyard disease detection
Organisation/Company Università degli Studi di Padova Department Math Num Research Field Computer science » Informatics Researcher Profile First Stage Researcher (R1) Positions PhD Positions Final date to receive applications 15 Apr 2026 - 00:00 (Europe/Brussels) Country Italy Type of Contract Temporary Job Status Full-time Offer Starting Date 1 Sep 2026 Is the job funded through the EU Research Framework Programme?

Horizon Europe - MSCA Marie Curie Grant Agreement Number  Is the Job related to staff position within a Research Infrastructure? No
Offer Description
PhD position (Position

E):
Optimizing images quality and deep learning methods for vineyard disease detection
Green Field Data :
IoRT Data Management and Analysis for Sustainable Agriculture

PUT, Poznan University of Technology, Poznan, Poland (18 Months)

Benefits

Exceptional benefits of MSCA PhD at a glance
International PhD training excellence
Interdisciplinary & multi sectoral research
Competitive MSCA salary & allowances
Global academic & industrial network
Non-academic secondments

Salary

Living Allowance
Mobility Allowance
Family Allowance

EUR 3822 EUR 710 EUR 660
Context
Viticulture is one of the most valuable agricultural sectors in Europe, but it faces increasing challenges from plant diseases that threaten both yield and quality. Early and reliable detection of vineyard diseases is crucial for ensuring sustainable production, reducing economic losses, and minimizing the use of chemical treatments. Current monitoring methods are often manual, time-consuming, and prone to inconsistencies, which limits their scalability and effectiveness.

Recent advances in robotics, imaging technologies, and artificial intelligence offer transformative opportunities for vineyard disease detection. High-resolution imaging combined with deep learning models enables precise recognition of disease symptoms at early stages, supporting more targeted interventions. However, image quality and data consistency remain significant bottlenecks: uncontrolled field conditions, variability in light, weather, and plant growth stages introduce noise that hampers reliable model performance.

This PhD project addresses these challenges by optimizing image acquisition and processing pipelines for vineyard monitoring, integrating advanced data augmentation techniques, and developing multimodal AI approaches that incorporate spatial variability factors. The outcomes will contribute not only to more robust and accurate disease detection systems but also to decision support tools that promote sustainable vineyard management. The research aligns with EU strategic priorities such as the Green Deal, Farm to Fork Strategy, and Digital EU Programme, bridging digital innovation with ecological sustainability in agriculture.
Objectives

Develop methods for early detection and classification of vineyard diseases using robotic platforms operating directly in the field.
Optimize and automate vineyard image acquisition and processing pipelines to ensure consistently high-quality datasets, addressing challenges of illumination, weather conditions, and plant variability.
Design and implement advanced data augmentation strategies to improve the robustness, generalization, and reliability of deep learning models for disease detection.
Incorporate spatial variability factors (soil, microclimate, plant development stage) into detection frameworks by leveraging multimodal approaches that integrate imaging, sensor data, and contextual information
Develop efficient data access and management mechanisms to support AI-driven queries and enhance the effectiveness of vineyard decision-support systems.

Work plan and task scheduling

Months 1–3:
Review state of the art on vineyard disease detection, imaging techniques, and deep learning approaches; identify key challenges in image quality, acquisition variability, and multimodal data integration. Define preliminary protocols for image collection in vineyards.
Months 3–6:
Acquire knowledge on vineyard pathophysiology and symptom expression of major diseases; establish initial image acquisition pipelines using robotic platforms (RGB, multispectral cameras). Begin…