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…

Iso-/anisohydric behavior in wine grapes may be a matter of soil moisture

There are claims that wine grape cultivars are either isohydric or anisohydric; the former maintaining, and the latter decreasing, their plant water status as soil moisture declines. However, available information is inconsistent. There are those that show an existence of a continuum in cultivar response to soil moisture rather than a distinct categorization. Others even show both behaviors in the same cultivar grown in different environments. In this study we investigated the behavior of 30 own rooted Vitis vinifera cultivars during successive drydown and rewatering cycles over two growing seasons in arid eastern Washington (<200 mm annual precipitation).

Study of the “Charentes terroir” for wine production of Merlot and Sauvignon: method, installation of the experimental device, first results

Cognac vineyard is mainly dedicated to brandy production. Within the vineyard restructuring context, one part is turned over wine varieties for wine production (about 1,500 ha planted from 1999 to 2005). Today, the new wine producers need technical references about qualitative potential of the « Charentes Terroir », varieties and adapted vineyard management.
In order to answer to this professional request, an observatory of 18 plots of Merlot and 12 plots of Sauvignon have been laid out since 2003 and 2004 on various kinds of pedoclimate.

Identification of aroma markers in amarone wines

Amarone is an Italian red wine produced in the Valpolicella area, in north-eastern Italy. Due to its elaboration with withered grapes, Amarone is a rather unique example of dry red wine. However, there is very limited data so far concerning the volatile composition of commercial Amarone wines, which also undergo a cask aging of 2-4 years before release. The present work aims at characterizing the aroma composition of Amarone and to elucidate the relationships between chemical composition and sensory characters. The analysis of 17 Amarone commercial wines from the same vintage (2015) was carried out by means of Gas Chromatography-Mass Spectrometry (GC-MS) and extracted by Solid Phase Extraction (SPE) and Solid Phase Micro Extraction (SPME). In addition, the sampled wines were subjected to a sensory evaluation in the form of sorting task.RESULTS: 70 volatile compounds were successfully identified and quantified, 30 of which were present in concentrations above their odor thresholds in all the samples. Using the odor activity value (OAV), the compounds that potentially contribute to Amarone perceived aroma are b-damascenone, ethyl and isoamyl acetate, ethyl esters (hexanoate, octanoate, butanoate, 3-methybutanoate), 4-ethyl guaiacol, 3-methylbutanoic acid, dimethyl sulfide (DMS), eugenol, massoia lactone, 1,4-cineol, TDN, cis/trans-whisky lactone. In certain samples, high OAVs were also observed for 4-ethyl phenol and 1,8-cineole.Results from the sorting task sensory analysis showed three clusters formed.

ACCUMULATION OF GRAPE METABOLITES IS DIFFERENTLY IMPACTED BY WATER DEFICIT AT THE BERRY AND PLANT LEVELS IN NEW FUNGUS DISEASE-TOLERANT GENOTYPES

The use of new fungus disease-tolerant varieties is a promising long-term solution to better manage chemical input in viticulture, but unfortunately little is known regarding these new hybrids fruit development and metabolites accumulation in front of abiotic stresses such as water deficit (WD). Thus, prior to the adoption of such varieties by the wine industry in Mediterranean regions, there is a need to consider their suitability to WD.

Decline of new vineyards in Southern Spain

In-season vineyard pest management relies on proper timing, selection, and application of products. Most of the research on pest management tends to focus on the influence of regional conditions on these aspects, with an emphasis on product timing and efficacy evaluation. One aspect that is not fully vetted in various vineyard regions is application (sprayer) technology. The purpose of this study was to determine the influence of regional conditions on sprayer performance in commercial wine grape vineyards in eastern Washington.