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…

Winemaking processes discrimination by using qNMR metabolomics

AIM: Metabolomics in food science has been increasingly used over the last twenty years. Among the tools used for wine, qNMR has emerged as a powerful tool to discern wines based on environmental factors such as geographical origin, grape variety and vintage (Gougeon et al., 2019a).

Temperature variability inside a wine production area and its effect on vine phenology and grape ripening. An example from the Saint-Emilion-Pomerol

AIM: the aim of this study was to develop a method for fine-scale temperature zoning. The effect of temperature variability on vine phenology and grape composition was assessed in the production area of Saint-Emilion

Implementing VIS-NIR spectroscopy as a rapid and non-intrusive technique for assessing anthocyanin and phenolic concentrations in Vitis vinifera L. Grenache whole grape berries

Anthocyanins and phenolic compounds play a crucial role in winemaking, contributing to the profile, flavor, color, texture, and stability of wine. Grape clusters, specifically Vitis vinifera L. cv. Grenache, were handpicked from a commercial vineyard sited in Tudelilla, La Rioja, Spain (42°18′ 52.26″, Long. -2°7′ 59.15″, Alt. 582 m) on five distinct dates from veraison to harvest during the 2015 season. Non-contact spectral measurements were conducted on intact grape berries using a VIS-NIR spectrometer operating in the 570 – 1000 nm spectral range under controlled laboratory conditions, positioned at a distance of 25 cm from the berries. The quantification of 16 anthocyanins and phenols in 120 grape clusters was performed using HPLC, established as the reference method for validating the spectral tool.

Proteomic profiling of grape berry presenting early loss of mesocarp cell vitality

From fruit set to ripening, the grape berry mesocarp experiences a wide range of dynamic physical, physiological, and biochemical changes, such as mesocarp cell death (MCD) and hydraulic isolation. The premature occurrence of such events is a characteristic of the Niagara Rosada (NR) variety, utilised as table grapes and winemaking. In our opinion, the onset of ripening would not cause MCD, but a down-regulation of respiratory enzymes during the early loss of cell viability, while maintaining membrane integrity. For this, we investigated three distinct developmental stages (green (E-L33), veraison (E-L35), and ripe (E-L39)) of NR berries by label-free proteomics, enzymatic respiratory activity and outer mesocarp imaging. Cell wall-modifying proteins were found to accumulate differently throughout ripening, while cytoplasmic membranes continue intact.

The Gibberellic-Acid Insensitive gene Vvgai1 impacts both vegetative growth and organogenesis rate in Vitis labruscana

Context and purpose of the study. As other perennial crops grapevine is facing the challenges of climate changes. One of the major issues is global warming and variations of the water budget.