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…

Le Pinot noir dans la zone AOC des “Colli Orientali del Friuli” (nord-est de l’Italie) : influence de la forme de taille sur les paramètres viticoles et œnologiques du raisin et du vin

Pinot noir is an interesting vat variety for the high quality products it provides in the most suitable areas. In France, the most important Pinot Noir growing areas are Burgundy, Champagne, Alsace and the Loire. In Italy, Pinot Noir is grown almost exclusively in the northern regions of Trentino-Alto Adige, Lombardy and Friuli-Venezia Giulia.

Use of hyperspectral data for assessing vineyard biophysical and quality parameters in northern Italy

A total of 39 study sites from 11 commercial vineyards located in two traditional growing areas of Northern Italy were identified for airborne hyperspectral acquisition in summer 2009 with the Aisa-EAGLE Airborne Hyperspectral Imaging Sensor.

Colour assessment of port wines using colorimetric and spectrophotometric methods

Colour is an important quality parameter in wines and is the result of a complex mixture of pigments
(including anthocyanins and their derivatives, quinones, xanthyllium compounds, etc.). Red wine colour changes over time as pigments react between themselves and with other wine macromolecules
(particularly polyphenols). During wine tasting, colour is normally assessed on the outer rim of the wine profile in a tilted glass, since most wines are too opaque to be analysed in the middle of the glass. Therefore, depending on the depth of observation considered, the perception of wine colour can be different.

Data deluge: Opportunities, challenges, and lessons of big data in a multidisciplinary project

Grapevine powdery mildew resistance is a key target for grape breeders and grape growers worldwide. The driver of the USDA-NIFA-SCRI VitisGen3 project is completing the pipeline from germplasm identification to QTL to candidate gene characterization to new cultivars to vineyards to consumers. This is a common thread across such projects internationally. We will discuss how our objectives and approaches leverage big data to advance this initiative, starting with genomics and computer vision phenotyping for gene discovery and genetic improvement. To manage and maintain resistances for long-term sustainability, growers will be trained through our nation-wide extension and outreach plan.

Chemical markers in wine related to low levels of yeast available nitrogen in the grape

Nitrogen is an important nutrient of yeast and its low content in grape must is a major cause for sluggish fermentations. To prevent problems during fermentation, a supplementation of the must with ammonium salts or more complex nitrogen mixtures is practiced in the cellar. However this correction seems to improve only partially the quality of wine [1]. In fact, yeast is using nitrogen in many of its metabolic pathways and depending of the sort of the nitrogen source (ammonium or amino acids) it produces different flavor active compounds. A limitation in amino acids can lead to a change in the metabolic pathways of yeast and consequently alter wine quality.