terclim by ICS banner
IVES 9 IVES Conference Series 9 A novel dataset and deep learning object detection benchmark for grapevine pest surveillance

A novel dataset and deep learning object detection benchmark for grapevine pest surveillance

Abstract

Flavescence dorée (FD) stands out as a significant grapevine disease with severe implications for vineyards. The American grapevine leafhopper (Scaphoideus titanus) serves as the primary vector, transmitting the pathogen that causes yield losses and elevated costs linked to uprooting and replanting. Another potential vector of FD is the mosaic leafhopper, Orientus ishidae, commonly found in agroecosystems. The current monitoring approach involves periodic human identification of chromotropic traps, a labor-intensive and time-consuming process.

Therefore, there is a compelling need to develop an automatic pest detection system, leveraging the recent progress in computer vision and deep learning techniques. However, the current progress in developing such a system is hindered by the lack of effective datasets to serve as ground-truth data for the training process.

To fill this gap, our study contributes a fully annotated dataset of S. titanus and Or. ishidae from yellow sticky traps. The dataset comprises more than 400 images, with 1000 identification per class. Guided by entomologists, the annotation task involved defining bounding boxes around relevant insects with corresponding class labels.

We trained and compared the performance of state-of-the-art object detection algorithms (YOLOv8 and Faster R-CNN). Pre-processing included automatic cropping to eliminate irrelevant background information and image enhancements to improve overall quality. Additionally, we tested the impact of altering image resolution, data augmentation, and single-class detection. Preliminary results achieved a high detection accuracy, with mAP@50 and F1-score above 90%, and mAP@50-95 around 70%, allowing a first deployment as an automatic annotation support tool.

DOI:

Publication date: June 14, 2024

Issue: Open GPB 2024

Type: Poster

Authors

Giorgio Checola1*, Paolo Sonego1, Valerio Mazzoni2, Franca Ghidoni3, Alberto Gelmetti3, Pietro Franceschi1

1 Research and Innovation Centre, Digital Agriculture Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy
2 Research and Innovation Centre, Plant Protection Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy
3 Technology Transfer Centre, Viticulture Unit, Fondazione Edmund Mach, S. Michele all’Adige, TN, Italy

Contact the author*

Keywords

insect detection, deep learning, smart pest monitoring, flavescence dorée, insect traps

Tags

IVES Conference Series | Open GPB | Open GPB 2024

Citation

Related articles…

Vintage influence on Grenache N, Syrah N and Mourvedre N in Côtes du Rhône (France)

Vintage is part of « terroir ». The aim of this work is to study, through vine and berry parameters, the effect of vintage on the three major red grape varieties in Côtes du Rhône : Grenache N, Syrah N and Mourvedre N. We first characterized vintages 1997 to 2003, highlighting similar features in grape development across the different cultivars since 2001 only.

Cross analytical and sensory differentiation of monovarietal white wines from four autochthonous grape varieties: focus on macromolecules

White wines contain macromolecules such as proteins, phenolic compounds and polysaccharides. On a sensory
level, these compounds contribute to the ‘mouthfeel’ that differentiates the white wines worldwide [1].

Methyl Jasmonate Versus Nano-Methyl Jasmonate: Effect On The Stilbene Content In Monastrell Variety

Stilbenes, a kind of non-flavonoid phenolic compounds, have been reported to be responsible for various beneficial effects. Their biological properties include antibacterial and antifungal effects, as well as cardioprotective, neuroprotective and anticancer actions (Guerrero et al. 2009).Several strategies can be used to increase stilbene content in grapes and one of them could be the use of elicitors such as methyl jasmonate. The use of this elicitor has been proven to be efficient in the production of secondary metabolites which increases the quality of wines, but its use also has some drawbacks such as its low water solubility, high volatility, and its expensive cost (Gil-Muñoz et al. 2021).

USING CHECK-ALL-THAT-APPLY (CATA) TO CATEGORIZE WINES: A DECISION-MAKING TOOL FOR WINE SELECTION

Bordeaux is the largest appellation vineyard in France. This contrasting vineyard with varied terroirs offers all styles of wine, resulting from the blending of several grape varieties. If these different profiles make the renown of Bordeaux wines, it can appear as a constraint when the aim is to study Bordeaux wines in their diversity. The selection of a representative sample can be performed by a sensory analysis carried out by trained panelists or by wine professionals, which can take several forms: consensus among experts, conventional descriptive analysis, typicality or quality evaluation. However, because of time, economic, and logistical constraints, these methods have limited applications. As an alternative to classical descriptive analysis, more intuitive methods that do not require training have been proposed recently to describe wines using an expert panel such as Napping, Free Choice or Flash Profiling, CATA or RATA.

Exploring the dynamic between yeast mannoproteins structure and wine stability

Mannoproteins are macromolecules found on the surface of yeast cells, composed of hyperbranched polysaccharide negatively charged chains by mannosyl-phosphate groups, fixed to a protein core. during the alcoholic fermentation and aging on lees, these mannoproteins are released from the yeast cell wall and become the main yeast-sourced polysaccharide in wine. due to their techno-functional properties, commercial preparations of mannoproteins can be used as additives to better assure tartaric and protein stability.