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…

Targeted and untargeted 1H-NMR analysis for sparkling wine’s authenticity

Studies on wineomics (wine’s metabolome) have increased considerably over the last two decades. Wine results from many environmental, human and biological factors leading to a specific metabolome for each terroir. NMR metabolomics is a particularly effective tool for studying the metabolome since it allows the rapid and simultaneous detection of major compounds from several chemical families.1 Quantitative NMR has already proven its effectiveness in monitoring the authenticity of still wines.

Detoxification capacities of heavy metals and pesticides by yeasts 

Winegrowing is still characterized by the extensive use of chemical fertilizers and plant protection products, despite strong recommendations to limit these practices. A part of these xenobiotics and metals are then found in grape juice and wine, causing a major health concern, as well as negatively affecting the fermentation process. In recent years, there has been renewed interest in non-Saccharomyces yeasts. These species have a wide phenotypic diversity, which would be exploited to broaden the aromatic palette of wines.

Integrated multiblock data analysis for improved understanding of grape maturity and vineyard site contributions to wine composition and sensory domains

Much research has sought to define the complex contribution of terroir (varieties x site x cultural practices) on wine composition. This investigation applied recent advances in chemometrics to determine relative contributions of vine growth, berry maturity and site mesoclimate to wine composition and sensory profiles of Shiraz and Cabernet Sauvignon for two vintages.

EUGENOL AS QUALITY MARKER OF WINES AND SPIRITS FROM HYBRID VINES: IMPACT OF DIFFERENT WINEMAKING AND DISTILLATION PROCESSES

Eugenol, widely spread in various plants notably cloves, basil and bay, was identified too in wines from hybrid grapes without contact with oak wood. This aromatic molecule presents a strong spicy note of clove and also antifongic properties. Eugenol was described as an endogenous compound of Baco blanc, from the grapes to the spirits of Armagnac area. Moreover, this compound is a chemical marker of Baco blanc products quality.
Influences of harvest time and different winemaking processes (settling, use of enzymatic preparations, lees content and stock time before distillation) on Baco blanc wine eugenol contents were explored using a two-levels full factorial Design of Experiments (DoEs).

Digitalization and valorization of the genotypic and phenotypic information retained within the FEM grapevine germplasm

The maintenance and valorization of genetic diversity is an undoubtable resource for the viticulture of the future, since the climate crisis is forcing us to think of new, more resilient varieties. For this reason, the grapevine germplasm of the Fondazione Edmund Mach has been continuously expanded in the last decade to a total of 3,120 accessions, whose trueness-to-type has been verified by means of the universal set of nine microsatellites. About two thirds are V. vinifera subsp. vinifera accessions, while the rest consists of naturalized and selected hybrids, V. vinifera subsp. sylvestris, and pure species. The genetic material has also been characterized over three consecutive years for ampelographic, vine development, and biotic stress response traits to be exploited for experimental purposes.