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…

La hiérarchisation des Coteaux du Languedoc: une application concrète du zonage vitivinicole

L’A.O.C. Coteaux du Languedoc est située dans le Sud de la France, dans la partie Ouest de la bordure méditerranéenne. Elle forme un vaste amphithéâtre largement ouvert sur la mer méditerranée. L’Appellation a été constituée en 1960 par le regroupement de 14 anciennes petites appellations d’origine représentant 55 communes éparpillées dans les départements de l’Aude, de l’Hérault et du Gard. Par la suite, plusieurs extensions successives ont conduit à un ensemble actuellement composé de 168 communes.

A new path for sustainable development. First results in a Venetian “bio-métaéthique company“ (Italy)

This paper will show the results of changes in income in a Veneto winery located in the municipality of Motta di Livenza (Treviso) in the North East of Italy, determined by the application of the “Holistic Universal Sustainability Charter” Metaethic 4.1CC “or” Sustainability Charter BIO-MétaÉthique 4.1CC “of GiESCO (Carbonneau, Cargnello, 2017).

Impact of smoke exposure on the chemical composition of grapes

Vineyard exposure to smoke can lead to grapes and wine which exhibit objectionable smoky and ashy aromas and flavours, more commonly known as ‘smoke taint’ [1, 2]. In the last decade, significant bushfires have occurred around the world, including near wine regions in Australia, Canada, South Africa and the USA, as a consequence of the warmer, drier conditions associated with climate change. Considerable research has subsequently been undertaken to determine the chemical, sensory and physiological consequences of grapevine exposure to smoke. The sensory attributes associated with smoke-tainted wine have been linked to the presence of several smoke-derived volatile phenols, such as guaiacols, syringols and cresols [2].

FLOW CYTOMETRY, A POWERFUL AND SUSTAINABLE METHOD WITH MULTIPLE APPLICATIONS IN ENOLOGY

Flow cytometry (FCM) is a powerful technique allowing the detection, characterization and quantification of microbial populations in different fields of application (medical environment, food industry, enology, etc.). Depending on the fluorescent markers and specific probes used, FCM provides information on the physiological state of the cell and allows the quantification of a microorganism of interest within a mixed population. For 15 years, the enological sector has shown growing interest in this technique, which is now used to determine the populations present (of interest or spoilage) and the physiological state of microorganisms at the different stages of winemaking.

Modification on grape phenolic and aromatic composition due to different leafroll virus infections

Viral diseases are reported to cause several detrimental effects on grapevine. Among them, leafroll, due to single or mixed infection of GLRaV1 and GLRaV3, and rugose wood, associated to GVA, are considered the most widespread and dangerous.