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…

The influence of different fertiliser applications and canopy management practices on the potassium content and pH of juice and wine of Vitis vinifera L. cvs. Cabernet-Sauvignon and Cabernet franc

In an attempt to reduce the pH of juice and wine, different fertiliser applications and canopy management practices were evaluated in South Africa in a field trial. Fertiliser treatments entailed no, CaSO4, Ca(OH)2, and MgSO4 fertilisation.

Preliminary steps of a protocol to isolate transcription factors bound to a specific DNA locus in grapevine using CRISPR-dCas9 system

Cis-acting regulatory elements are DNA sequences that can be bound by transcription factors to regulate the expression of genes in a condition-dependent and tissue-specific way. It is nowadays possible to search for DNA motives and sequences that a given transcription factor is binding or at least can, but it is still hard to have a glance at all the transcription factors that are contemporaneously located at the same locus. Inspired by an existing technique that uses the CRISPR-Cas system in mammal cells, we are trying to develop a protocol to study such regulation in Vitis vinifera. Using the highly sequence-specific binding capacity of a catalytically inactive Cas9 protein (dCas9), our idea is to set up a system to target a desired sequence and precipitate all the crosslinked proteins and distantly interacting chromatin at this locus and analyze them.

Atypical ageing defect in Pinot Blanc wines: influence of the grapevine production management.

Atypical ageing (ATA) is a wine aroma fault occurring in white wines characterised by an early loss of varietal aroma as well as nuances of wet mop, acacia blossom, shoe polish and dirty rag among others. 2-aminoacetophenone (2AAP) – a degradation product of indole-3-acetic acid (IAA) – has been described as the major odour-active compound and chemical marker responsible for this off-flavour. Depending on the aroma intensity of wines, its odour threshold varies from 0.5 to 10.5 μg/L. It seems that a stress reaction in the vineyard triggered by climatic, pedological and viticultural factors can ultimately cause ATA development in wines and therefore shorten their shelf-life.

Extreme canopy management for vineyard adaptation to climate change: is it a good idea?

Climate change constitutes an enormous challenge for humankind and for all human activities, viticulture not being an exception. Long-term strategic changes are probably needed the most, but growers also need to deal with short-term changes: summers that are getting progressively warmer, earlier harvest dates and higher pH in musts and wines. In the last 10-15 years, a relevant corpus of research is being developed worldwide in order to evaluate to which extent extreme canopy management operations, aimed at reducing leaf area and, thus, limiting the source to sink ratio, could be useful to delay ripening. Although extreme canopy management can result in relevant delays in harvest dates, longer term studies, as well as detailed analysis of their implications on carbohydrate reserves, bud fertility and future yield are desirable before these practices can be recommended.

Taking advantage of difficulties. Variable rate application based on canopy maps to achieve a sustainable crop

Aim: The aim of this work was to evaluate the use of Variable Rate Application technologies based on prescription maps in commercial vineyards with large intra-parcel variability to achieve a more sustainable distribution of Plant Protection Products (PPP)