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…

Shading grapevines with dynamic agrivoltaics address the challenge of early ripening and wine quality related with climate change

Context and purpose of the study. Climate change accelerates grapevine’s phenology, advancing harvests by 2–3 weeks over the past 40 years negatively affecting wine style due to a lack of acidity and too much alcohol.

Le réseau français des partenaires de la sélection vigne : un dispositif unique au monde au service de la sauvegarde du patrimoine variétal

The French vine selection partners network is currently made up of 40 regional partners, grouped around IFV (French Institute for Vine and Wine) and INRAE (national research institute for agriculture and environment), whose missions are preservation, selection, and innovation of our varietal diversity. The originality of this device is based on a 3-level organisation: – varietal diversity preservation, with the world reference: the INRAE’s vine genetics resources centre of Vassal-Montpellier (Marseillan, France), the world’s largest ampelographic collection, which includes nearly 6 000 accessions of cultivated Vitis vinifera from 54 countries, as well as rootstocks, interspecific hybrids, wild vines (lambrusques) and wild American and Asian species.

The effect of ecological conditions on the germination of pollen, fecundation and yield of some grapevine cultivars in Skopje region, Republic of Macedonia

The ecological conditions (climatic factors and soil) during the whole year, and especially before flowering and during the time of flowering, have a great influence on the functional ability of pollen, the pollination, the fecundation and the yielding potential of the cultivars of grapevine.

Techniques to study graft union formation in grapevine

Grapevines are grown grafted in most viticultural regions. Grapevine rootstocks are either hybrids or pure species of different American Vitis spp. (particularly V. berlandieri, V. rupestris and V. riparia), which were primarily used to provide root resistance to the insect pest Phylloxera. In addition to Phylloxera resistance, grapevine rootstocks were also selected in relation their resistance to various abiotic stress conditions. Future rootstocks should have the potential to adapt viticulture to climate change without changing the characteristics of the harvested product. However, high grafting success rates are an essential prerequisite to be able to use them with all the varieties. The objective of this work is to develop quantitative techniques to characterize graft union formation in grapevine.

Effects of severe shoot trimming at different phenological stages on the composition of Merlot grapes

High concentration of sugars in grapes and alcohols in wines is one of the consequences of climate change on viticulture production in several wine regions. One of the options to alleviate this potential problem is to perform severe shoot trimming of the vines to limit the production of carbohydrates. Two different studies were performed in order to investigate the effects of severe shoot trimming on the composition of Merlot grapes; in a first study severe shoot trimming was performed at three different phenological stages (at berry set, at the beginning of veraison and at the end of veraison), while in a second study two trimming treatments (standard shoot trimming and severe shoot trimming performed at the end of veraison) were combined with two shoot densities in order to evaluate the relative impact of these treatments on Merlot grape composition.