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…

Glutathione content evolution during spontaneous alcoholic fermentations of Sangiovese grapes

Glutathione is a tripeptide (γ-Glu-Cys-Gly), which can occur in grapes, in must and in wine prevalently in the reduced form as well as in the oxidized form as glutathione disulfide. The importance of the reduced form of glutathione lies in its antioxidant activity. In must, it limits browning by reducing o-quinones produced by polyphenol oxidase activity on hydroxycinnamic acids; in wine, it exerts a protective effect on various aromatic compounds. Glutathione concentration in wine is lower than in grape juice and variable as it depends on several factors, ranging from the native content of grapes to winemaking technique.

Possible methods of adaptation to the effects of climate change in the Tokaj Wine Region 

Viticulture’s adaptation to the harmful effects of climate change is globally the biggest challenge of the near future. Short, extremely intensive rainfalls and longer periods of drought are getting more frequent in the Tokaj Wine Region, where the majority of the vineyards are cultivated on steep slopes. Hence, erosion has high risk, especially when combined with the loess-based soils on about ten percent of the region. The environmentally beneficial cover crop and mulch usage can effectively reduce the risk of erosion, according to research done by the Tokaj Wine Region Research Institute of Viticulture and Oenology.

Multi-trait selection in ancient grapevine varieties

The selection of ancient grapevine varieties aims to achieve genetic gains in several important traits that can make the variety more interesting for the objectives of the producers. Traditionally, yield and quality traits of the must have been considered for selection, but many others can be taken into account. Linear mixed models are fitted to the data to predict the empirical best linear unbiased predictors (EBLUPs) of genotypic effects for each evaluated trait, which will be the basis for selection.

Physiological means to curb the enthusiasm of viruses from infecting grapevines  

The two most deadly viruses infecting and threatening the productivity of grapevines worldwide are leafroll and red blotch viruses. There is no cure for viral diseases other than roguing the symptomatic vines and replacing them with certified vines derived from clean, virus-tested stocks.
Given that phloem plays a central role in virus infection, this study aimed to purge the virus by girdling the phloem of leafroll-infected vines at different phenological stages of infected grapevines. Phloem-girdling was performed on canes at veraison to varying regions between the proximal and distal clusters.

A pragmatic modeling approach to assessing vine water status

Climate change scenarios suggest an increase in temperatures and an intensification of summer drought. Measuring seasonal plant water status is an essential step in choosing appropriate adaptations to ensure yields and quality of agricultural produce. The water status of grapevines is known to be a key factor for yield, maturity of grapes and wine quality. Several techniques exist to measure the water status of soil and plants, but stem water potential proved to be a simple and precise tool for different plant species.