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…

Effect of intra‐vineyard ripeness variation on the efficiency of commercial enzymes on berry cell wall deconstruction under winemaking conditions

Intra-vineyard variation grape berry ripening occurs within bunches, between bunches on the same vine and between vines. Although it is assumed that such variation also occurs at the grape berry cell wall level, no study to data has investigated in any depth. Here we have used a intra-vineyard panel design to investigate pooled bunches from six vines (per panel) in the context of a winemaking scenario. The dissected vineyard was harvested by separate panels, where each panel was then subjected to a standard winemaking procedure with or without the addition of three different enzyme preparations for maceration.

DETERMINATION OF FREE AMINO ACIDS, AMINO ACID POTENTIAL AND PROTEASE ACTIVITY IN THE LEES AND STILL WINES OF CHAMPAGNE

Prior to winemaking, organic or mineral nitrogen compound concentrations are usually measured in the vineyard and in grape musts. These indicators facilitate vine cultivation decisions, usually through yield or vigor. During vinification, yeast and bacteria metabolize nitrogen compounds in the musts in order to generate biomass. After fermentation, the microorganisms rerelease a part of this nitrogen as soluble compounds into the wines. Another part remains bound in the lees and can be lost during racking. The must’s natural nitrogen quantities, additional supplements during fermentation, and lees contact management enhance the release of nitrogen compounds to the wines. During ageing these nitrogen compounds – primarily the amino acids – are implicated in the generation of odorous compounds such as heterocycles(1).

Enological characterization of mold resistant varieties grown in the Trentino Alpine Region

Among the different strategies used in vine growing to fight against mold diseases, it can be pointed out the hybridation of traditional grape varieties with others, presenting a genetic resistance to pathogen attack. The research in this field has been encouraged in recent years due to the increased concern about human safety and environmental pollution linked to the use of agrochemicals. This approach allows to limit the number of treatments and the type of active compounds used in vine management. The environment determines the pressure degree of the diseases on vines and the biologic response of the plant to their attack.

Zoning mountain landscapes for a valorisation of high identity products

Mountain agriculture is made difficult by the geomorphological complexity of the territory. This is especially true for viticulture: over the centuries the work of men in such a difficult environment

Wines empirical perception and growers management practices in the Anjou Villages Brissac vineyard (France)

The concept of viticultural terroir includes soil, sub-soil, and climatic factors but also many management viticultural and oenological practices which are chosen according to know-how of the winegrowers.