terclim by ICS banner
IVES 9 IVES Conference Series 9 GiESCO 9 Implementation of a deep learning-based approach for detecting and localising automatically grapevine leaves with downy mildew symptoms

Implementation of a deep learning-based approach for detecting and localising automatically grapevine leaves with downy mildew symptoms


Context and purpose of the study – Grapevine downy mildew is a disease of foliage caused by Oomycete Plasmopara viticola an endoparasite that develops inside grapevine organs and can infect virtually every green organ. Downy mildew is one of the most destructive diseases in wine-growing regions, drastically reducing yield and fruit quality. Traditional manual disease detection relies on farm experts. Human field scouting has been widely used for monitoring the disease progress, however, is costly, laborious, subjective, and often imprecise. In this sense, computer vision technologies and artificial intelligence provide a suitable alternative to improve the current disease detection techniques. Therefore, this study aims to validate a deep learning-based approach for detecting and localising automatically leaves with downy mildew symptoms.

Material and methods – Fourteen commercial blocks (different grapevine cultivars) located in northern Spain were assessed to generate a comprehensive dataset for validating the deep learning algorithm. All analysed blocks presented downy mildew symptoms with different levels of intensity. RGB Images of the canopy were taken manually using a conventional camera. The images were acquired during different daily hours and under contrasting light conditions. YOLOv4 (You Only Look Once) was the deep learning algorithm analysed in this study. YOLOv4 model was trained using a heterogeneous dataset populated by RGB images obtained from the different vineyard blocks under different conditions to increase the robustness of the model. The RGB images were carefully labelled manually by an expert, selecting leaves with visible downy mildew symptoms. The labelled images were divided into two datasets, using 80% for training and validating and 20% for testing. The images used for training were divided into 1500 x 1500 pixel sub-images obtaining 15 sub-images per image and the sub-images were resized to 640×640 pixels. Data processing and deep learning modelling were performed with Python programming language and the Darknet neural network framework. The metrics used to evaluate the model were mean Average Precision (mAP), F1-score and Intersection over Union (IoU).

Results – The results of YOLOv4 for detecting leaves with downy mildew symptoms are promising. In the testing process applied on full images (2560×1728 pixels), the model presented a mAP of 67%, an F1-score of 0.69 and an IoU of 62%. When the number of real infected leaves (labelled by an expert) was compared with the predicted number of infected leaves the model reached a determination coefficient R2 of 0.93.  The accuracy of the method to determine the number of infected leaves was similar across the whole range of infections. This indicates that the model fits appropriately to all conditions tested. Also, the analysis of the localization indicates that leaves with apparent symptoms were detected correctly by the model.


Publication date: July 5, 2023

Issue: GiESCO 2023

Type: Poster


Carlos POBLETE-ECHEVERRÍA1,2, Inés HERNÁNDEZ1,2, Salvador GUTIÉRREZ3, Rubén ÍÑIGUEZ1,2, Ignacio BARRIO1,2 and Javier TARDAGUILA1,2*

1Televitis Research Group, University of La Rioja, 26006 Logroño, Spain
2Institute of Grapevine and Wine Sciences (University of La Rioja, Consejo Superior de Investigaciones Científicas, Gobierno de La Rioja), 26007 Logroño, Spain
3Department of Computer Science and Artificial Intelligence (DECSAI), Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada (UGR), 18071 Granada, Spain

Contact the author*


artificial intelligence, deep learning, disease detection, precision viticulture, YOLOv4


GiESCO | GIESCO 2023 | IVES Conference Series


Related articles…

Effect of foliar application of Ca, Si and their combination on grape volatile composition

Calcium (Ca) is an important nutrient for plants which plays key signaling and structural roles. It has been observed that exogenous Ca application favors the pectin accumulation and inhibition of polygalacturonase enzymes, minimizing fruit spoilage. Silicon (Si) is a non-essential element which has been found to be beneficial for improving crop yield and quality, as well as plant tolerance to diverse abiotic and biotic stress factors. The effect of Si supply to grapevine has been assessed in few investigations, which reported positive changes in grape quality and must composition.

Late winter pruning induces a maturity delay under temperature-increased conditions in cv. Merlot from Chile

Chile is considered vulnerable to climate change; and these phenomena affect several mechanisms in the grape physiology and quality. The global temperature increase affects sugar contents, organic acids, and phenolic compounds in grapes, producing an imbalance maturity. In this sense, an alternative to reduce the impact is to perform pruning after vine budburst, known as “Late Pruning” (LP).

Effect of two water deficit regimes on the agronomic response of 12 grapevine varieties cultivated in a semi-arid climate

The Mediterranean basin is one of the most vulnerable regions to Climate Change effects. According to unanimous forecasts, the vineyards of Castilla-La Mancha will be among the most adversely affected by rising temperatures and water scarcity during the vine’s vegetative period. One potential strategy to mitigate the negative impacts of these changes involves the identification of grapevine varieties with superior water use efficiency, while ensuring satisfactory yields and grape quality.

Influence of irrigation frequency on berry phenolic composition of red grape varieties cultivated in four spanish wine-growing regions

The global warming phenomenon involves the frequency of extreme meteorological events accompanied by a change in rainfall distribution. Irrigation frequency (IF) affects the spatial and temporal soil water distribution but its effects on the phenolic composition of the grape have been scarcely studied. The aim of this work was to evaluate the effects of four deficit irrigation frequencies of 30 % ETo: one irrigation per day (T01), two irrigations per week (T03), one irrigation per week (T07) and one irrigation every two weeks (T15) on berry phenolic composition at harvest.

Response of red grape varieties irrigated during the summer to water availability at the end of winter in four Spanish wine-growing regions: berry phenolic composition

Water availability is the most limiting factor for vineyard productivity under Mediterranean conditions. Due to the effects caused by the current climate change, wine-growing regions may face serious soil moisture conservation problems, due to the lower water retention capacity of the soil and higher soil irradiation. The aim of this work was to evaluate the effects of soil recharge irrigation in pre-sprouting and summer irrigation every week (30 % ETo) from the pea size state until the end of ripening (RP) compared to exclusively summer irrigation every week (R) in the same way that RP, on berry phenolic composition at harvest.