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…

Organic and biodynamic viticulture affect soil quality and soil microbial diversity

The production of organically grown crops developed exponentially in the last few decades based on consumer demands for healthy food

Soil and topography effects on water status and must composition of chardonnay in burgundy & a mini meta‐analysis of the δ 13C/water potentials correlation

The measurement of carbon isotopic discrimination in grape sugars 13 at harvest (δ C) is an integrated assessment of water status during ripening.

CONSUMER PERCEPTION OF INTERSPECIFIC HYBRID RED WINE COLOR IN RELATION TO ANTHOCYANIN PROFILE AND CHEMICAL COLOR PARAMETERS

Interspecific hybrid winegrapes are of growing interest in the context of climate change based on their disease resistance and cold hardiness. In addition to a need for increased understanding of their chemical composition, there is little empirical evidence on the consumer perception of non-vinifera wine. Phenolic compounds, and particularly color, play an important organoleptic and quality determination role in wine, but can vary significantly in interspecific hybrid wines compared to wines produced from Vitis vinifera cultivars [1, 2, 3]. Anecdotally, the variation in anthocyanin species, interactions, and concentrations in interspecific hybrids could result in a variance from“vinifera-like” wine color.

Adsorption of tetraconazole by organic residues and vineyard organically-amended soils 

Spain is the country with the largest wine-producing area in the EU and its productivity is largely controlled applying fungicides. However, residues of these compounds can move and contaminate surface and groundwater. The objective of this work was to evaluate the capacity of bioadsorbents from different origin to adsorb and immobilize tetraconazole by themselves or when applied as organic soil amendment, and to prevent soil and water contamination by this fungicide. The adsorption of tetraconazole by 3 organic residues: spent mushroom substrate (SMS), green compost (GC) and vine pruning sawdust (VP), as well as by vineyard soils unamended and amended individually with these residues at 1.5% (w/w) was evaluated using the batch equilibrium technique.

A population genetic study of Vitis vinifera L. subsp. sylvestris Gmelin based on 3.000 individuals from 20 countries

Until the 19th century, the wild form of cultivated grapevines (vitis vinifera l. subsp. sylvestris gmelin, v. sylvestris) was ubiquitous in many european and west asian regions. However, many factors like deforestation, the intensification of agriculture, or the introduction of several pests and pathogens decimated its presence in these growing sites, and natural populations are now mostly restricted to river-bank forests and creeks with specific soil and climate conditions. in fact, v. sylvestris is now considered an endangered subspecies that is protected by law in many european countries to prevent its loss.