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…

The 1000 grapevine genomes project: Cataloguing Australia’s grapevine germplasm

Grapevine cultivars can be unequivocally typed by both physical differences (ampelography) and genetic tests. However due to their very similar characteristics, the identification of clones within a cultivar relies on the accurate tracing of supply records to the point of origin. Such records are not always available or reliable, particularly for older accessions. Whole genome sequencing (WGS) provides the most highly detailed methodology for defining grapevine cultivars and more importantly, this can be extended to differentiating clones within those cultivars.

Ceramic imprint in wine: influence of hydraulic ratio on ceramic dissolution and wine pH in amphorae systems

This interaction is primarily due to an acidic attack on the ceramic by the wine. It results in (1) the dissolution of the ceramic into the wine and the release of a wide variety of elements; and (2) an increase of the wine pH. The extent of these effects depends on the mineralogical and chemical composition of the ceramic, as well as the hydraulic ratio of the ceramic-wine system (the term hydraulic ratio (ρ) defines here the volume of wine over the surface area of the ceramic in contact with the wine).

Reconfiguring wine prescription : from traditional critics to digital social networks

The integration of digital social networks (DSN) has profoundly transformed communication practices within the wine industry, reorganizing the dynamics of prescription and marketing.

Exploring the prevalence of esca-induced leaf symptoms in French vineyards and the role of climate: a national scale analysis

Esca, a severe trunk disease affecting vineyards, is caused by fungal pathogens that induce wood necrosis and decay, leaf symptoms, yield losses, and potentially a rapid death of the vine. The prevalence of this disease varies across years, regions, cultivars, and plot ages. Despite its significance in understanding and predicting dieback risk in different vineyards, the role of climate in trunk diseases remains a relatively unexplored research area. While some studies have demonstrated the impact of certain climatic conditions on the prevalence of the disease, they often focus on a limited number of plots and yield conflicting results.We conducted a statistical analysis, using a Bayesian approach on a national database comprising prevalence data of esca from over 500 different plots in France, spanning the years 2003 to 2022 and encompassing various cultivars.

Metabolomic insights into wine’s sensory identity: unveiling climate-driven changes in aroma composition

Wine, a sensitive and intricate agricultural product, is being affected by climate change, which accelerates grapevine phenological stages and alters grape composition and ripening. This influences the synthesis of key aroma compounds, shaping wine’s sensory attributes [1]. The complex aroma profile, resulting from compound interactions, presents a metabolomics challenge to identify these indicators and their environmental change responses, which is being addressed using diverse analytical techniques.