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…

Cluster trait prediction using hyperspectral signatures in a population of 221 Riesling clones

Cluster architecture in grapevine plays a critical role in influencing bunch microclimate, thus quality traits, including sugar content, phenolic composition, and disease susceptibility.

The effectiveness of proximal remote sensors in plant water status evaluation of grapevine

Extensive studies have been conducted on grapevine responses to water deficit, but these responses are difficult to generalise since numerous factors can influence the response(s), including genotype, developmental stage, soil, climate, and season.

Mean polymerization degree of proanthocyanidins of grape seeds, skins and wines from Agiorgitiko (cv. Vitis vinifera): Differences among vintages

Grape phenolic compounds are very important constituents of red wine because, in addition to their antioxidant properties, they contribute to color, astringency and bitterness, oxidation reactions, interactions with proteins and ageing behavior of wines. The aim of our study was to assess the structural characteristics of grape and wine proanthocyanidins of Agiorgitiko variety and to evaluate the influence of the vintage year. Twelve vineyard locations were designated in the Nemea wine region. For three consecutive years (2012-2014), the grapes were harvested at technological maturity and the method of phloroglucinolysis was employed to determine the mean degree of polymerization (mDP) and subunit composition of the samples.

Identification of several glycosidic aroma precursors in six varieties of winemaking grapes and assessment of their aroma potential by acid hydrolysis

In winemaking grapes, it is known that most aroma compounds are present as non-volatile precursors, such as glycosidic precursors. In fact, there is strong evidence supporting the connection between the content of aroma precursors and the aromatic quality of wine [1]. Acid hydrolysis is preferred to reveal the aroma potential of winemaking grapes, as it predicts more accurately the chemical rearrangements occurring during fermentation in acidic environments [2]. In this study, a method involving a fast fermentation followed by acid hydrolysis at 75ºC was used to evaluate the accumulation of aroma compounds over time in fractions obtained from six different varieties of winemaking grapes.

Dispersive liquid-liquid microextraction for the quantification of terpens in wines

In a highly competitive worldwide market, a current challenge for the beverage sector is to diversify the range of products and to offer wines and spirits with typicity and character.

During alcoholic fermentation, wine yeasts generate a large variety of volatile metabolites, including acetate esters, ethyl fatty acid esters, higher alcohols, volatile fatty acids and volatile sulfur compounds that contribute to the aroma profile of wine. These molecules, refered as fermentative aromas, are the most abundant volatile compounds synthetized by yeasts and the metabolic pathways involved in their formation have been well characterized. Furthermore, other molecules with a major organoleptic impact may be produced during wine fermentation including terpene derivatives. However, little information is available on the contribution of yeasts to the formation of these molecules, in particular on their ability to synthethise de novo the terpens derivatives or to produce hydrolytic enzymes involved in the release of varietal precursors.