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…

An effective approach to mitigating ochratoxin A (OTA) levels in wine with minor impact on wine quality

OTA occurrence in wine is well-documented, with higher levels typically found in red (< 0.01-7.63 μg/l), followed by rose (0.01-2.40 μg/l) and white wine (<0.01-1.72 μg/l). Incidence rates are nOTAble, with studies showing OTA present in 53% of 521 red wines, 69% of 98 rose, and 61% of 301 white wines analysed. In europe, wine is estimated to be the second source of OTA intake after cereals. Since 2006, the maximum allowable limit for OTA in wine is 2 μg/l, according to regulation (ec) no. 1881/2006.

Pure wine vs natural wine

S’il n’existe pas de réglementation officielle, la démarche des vins naturels prône un retour aux pratiques dites ancestrales préconisant notamment un mode d’élaboration des vins utilisant le moins d’intrants possible. Le seul autorisé reste l’anhydride sulfureux (SO2) à des doses quatre à cinq fois moins importantes que pour les vins dits conventionnels. Ce désir de renouer avec

“Garrigues”, part of the mediterranean vine terroirs

Les paysages viticoles méditerranéens présentent une originalité qui ne se retrouve nulle part ailleurs : ils associent des garrigues très odoriférantes à des parcelles de vignes souvent qualitatives. La connaissance empirique des vins du Languedoc par leurs dégustateurs a conduit la Chambre d’Agriculture de l’Hérault à supposer que les arômes de la garrigue environnante peuvent se retrouver dans les vins (arômes de ciste, de genévrier, … )

From genes to vineyards: system biology and new breeding technologies for water stress tolerance in grapevines

One of the major challenges for food security and sovereignty is to produce stress-tolerant plants without introducing foreign DNA, because the legislative process, that bans transgenics, challenges us to find new solutions for producing plants that can survive the drought. To achieve this goal, we need to identify genes that can be modified to improve stress tolerance in plants. In this work, we present an online tool for exploring the transcriptome of grapevines under water stress, which is one of the most important abiotic stresses affecting viticulture. The tool is based on a comprehensive collection of rna-seq data from 997 experiments, covering four different tissues (leaf, root, berry, and shoot), various levels of water stress, and diverse genetic backgrounds (cultivars and rootstocks) with different levels of tolerance to water stress.

Visualization of wine origin, quality level and terroir by the landscape

The communication of the aims of a viticulture under the premise of terroir is presently discussed in a lot of wine-growing regions around the world. To encourage this discussion the differences in knowledge, understanding, and preference concerning wine and landscape should be regarded more closely: the wine should be perceived as a representative of its region and one of the most characteristic features of a region is the landscape.