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 chemical composition of disease resistant hybrid grape cultivars and its impact on wine quality: an exploratory enquiry into sustainable wines

Disease resistant hybrid grape cultivars are now allowed in a number of EU wine PDOs, and are also accepted in a number of countries outside the EU. There is increasing interest in diseases resistant hybrid grape cultivars (RHGCs) because they allow for the production of healthy, high quality grapes with limited use of pesticides and the associated environmental and public health

Implementation of hyperspectral image analysis for evaluating table grape quality on bunch and berry level

Typically, subjective, and visual methods are used by grape growers to assess harvest maturity. These methods may not accurately represent the maturity of an entire vineyard – especially if extensive and representative sampling was not used. New technologies have been investigated for improved harvest management decisions. Spectroscopy methods utilizing the near-infrared region of the light spectrum is one such technology investigated as an alternative to classic methods and particularly the application of hyperspectral imaging (HSI) has recently gained attention in research. HIS is a spectroscopic technique that obtains hundreds of images at different wavelengths collecting spectral data for each pixel in the sample i.e., providing both spectral and spatial data.

The influence of native flora on Argentine white terroir cv. Torrontes Riojano

The main objective of this paper is to establish considerable differences between wines from three wine areas or terroir, made with cv Torrontes Riojano.

Study of the volatile aroma profile of five Italian grape varieties submitted to controlled postharvest withering

Wines made with grapes submitted to postharvest dehydration are often referred to as “passito” or “straw wines.” This distinct style of winemaking consists of a process of water loss that allows the berries to undergo a mild water stress and senescence process [1].

Do natural wines differ from conventionally-produced wines?

In recent years, consumer awareness for consuming healthy and environmental sustainability products has considerably increased [1]. In an ever-changing and highly competitive environment such as the wine sector, production of wines without sulfites, or biodynamic, organic or vegan wines, has experienced an important increase to meet the new needs of consumers [2,3]. Beyond these categories of regulated products, a new concept has emerged: natural wines (NW), for which there is not an established definition or legal regulation. Rather, producers have a personal idea of naturalness under the premise of applying minimal intervention from grape to wine production [4]. In this context, it is hypothesized that self-defined natural wines are different from conventional wines (CW) in their sensory and chemical profile. The predicament of natural wine is based on anecdotic declarations and assumes that minimal intervention guarantees the production of wines with organoleptic properties able to express the “terroir” and thus promote wine diversity, plurality and sensory typicity against the risk of standardization of CW.